Wheat harvesting loss monitoring system and method

Abstract

The invention relates to the technical field of grain loss monitoring, in particular to a wheat harvesting loss monitoring system and method, and the system comprises an unmanned aerial vehicle monitoring module, a multi-mode sensor array, a signal processing module, an intelligent control module and a man-machine interaction terminal. The unmanned aerial vehicle monitoring module is used for scanning a field parcel, identifying a lodging area and a high-density area and generating a loss risk thermodynamic diagram, the loss risk thermodynamic diagram is imported into a harvester navigation system, and the high-risk area is harvested preferentially; the multi-mode sensor array comprises a piezoelectric grain sensor and an image sensor; the piezoelectric type grain sensor is mounted at the tail part of the cleaning sieve, adopts a vibration isolation device (such as a rubber pad) to reduce mechanical vibration interference, and detects a grain impact signal through a piezoelectric element; the image sensor is arranged on a header and a threshing outlet and is used for identifying missed cutting of wheat ears and damaged grains based on a deep learning algorithm; the intelligent control module comprises a loss rate calculation unit and a parameter optimization unit; the precision and the real-time performance are improved.

Description

Technical Field

[0001] This invention relates to the technical field of grain loss monitoring, and in particular to a wheat harvest loss monitoring system and method. Background Technology

[0002] During mechanized wheat harvesting, grain loss mainly includes incomplete threshing, cleaning losses, missed cuts by the header, and grain breakage. Current technologies for loss monitoring largely rely on manual sampling or single-sensor detection, which suffers from poor real-time performance, low accuracy, and the inability to dynamically adjust operating parameters. Furthermore, the complex wheat harvesting environment (vibration, dust, and interference from various impurities) makes traditional sensors susceptible to interference, leading to significant monitoring errors. Therefore, a high-precision, real-time monitoring system with automatic control is urgently needed. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a wheat harvest loss monitoring system and method.

[0004] The present invention provides a wheat harvest loss monitoring system, comprising an unmanned aerial vehicle (UAV) monitoring module, a multimodal sensor array, a signal processing module, an intelligent control module, and a human-machine interaction terminal; The drone monitoring module is used to scan fields, identify lodged areas and high-density areas, and generate a "loss risk heat map". The loss risk heat map is imported into the harvester navigation system to prioritize the harvesting of high-risk areas. The multimodal sensor array includes a piezoelectric grain sensor and an image sensor; Piezoelectric grain sensor: Installed at the tail of the cleaning screen, it uses vibration isolation devices (such as rubber pads) to reduce mechanical vibration interference and detects grain impact signals through piezoelectric elements; Image sensors: deployed at the header and threshing exit, using deep learning algorithms to identify missed ears of wheat and damaged grains; The intelligent control module includes a loss rate calculation unit and a parameter optimization unit; The loss rate calculation unit calculates the loss rate for each sampling area for rice, wheat, and corn grain harvesting according to the following formula:

[0005] In the formula: S—loss rate, in terms of %; W—grain loss of rice (corn, wheat) in the sampling area, in terms of grams; M—grain yield of rice (corn, wheat) per unit area, in terms of kilograms per mu; L is taken as 2 meters or the working width of a commonly used combine harvester in the local area, in terms of meters; The parameter optimization unit is used to establish a mapping model between the loss rate and operating parameters (such as threshing gap, sieve opening, and fan air volume), and automatically adjusts the harvester parameters through a PID controller; The human-machine interface terminal displays real-time loss rate, operational parameter suggestions, and alarm information, and supports data export and analysis. The drone monitoring module scans the field, identifies lodged areas and high-density areas, and generates a "loss risk heat map." This heat map is imported into the harvester navigation system to prioritize harvesting high-risk areas. A piezoelectric grain sensor is installed at the tail of the cleaning screen, using vibration isolation devices (such as rubber pads) to reduce mechanical vibration interference and detecting grain impact signals through piezoelectric elements. Image sensors are deployed at the header and threshing outlet, using deep learning algorithms to identify missed ears of wheat and damaged grains. The loss rate calculation unit calculates the loss rate for each sampling area according to a formula. The parameter optimization unit establishes a mapping model between the loss rate and operational parameters (such as threshing gap, screen opening, and fan airflow), automatically adjusting harvester parameters through a PID controller. The human-machine interface terminal displays real-time loss rate, operational parameter suggestions, and alarm information, and supports data export and analysis, achieving a high-precision, real-time, and automatically adjustable monitoring system.

[0006] Preferably, the multimodal sensor array further includes a header height sensor, a reel speed sensor, and a travel speed sensor, which work together to collect operating parameters in real time.

[0007] Preferably, the drone monitoring module includes a drone body, propellers, support rods, a base plate, and a housing. The drone body is equipped with multiple sets of propellers, and two sets of support rods are located at the bottom of the drone body. Each set of support rods has a base plate at its bottom. The housing is installed at the bottom of the drone body and also includes a crop lodging monitoring component. The crop lodging monitoring component is installed inside the housing. In use, the multiple sets of propellers are operated to rotate, causing the drone body to take off. During takeoff, the crop lodging monitoring component automatically extends to the outside of the housing and identifies lodged areas and high-density areas, generating a heat map of loss risk. After use, the drone body lands, and the base plate and support rods work together to provide stable support for the drone body. Simultaneously, the crop lodging monitoring component automatically enters the housing for storage, improving safety and convenience.

[0008] Preferably, the crop lodging monitoring component includes a rotating shaft, a multispectral camera, a lidar, gears, optical bars, a sliding rod, a lifting plate, a rack, a spring, a limiting plate, and a support plate. The support plate is rotatably mounted on the inner bottom of the housing with the cooperation of the rotating shaft. The support plate seals the bottom of the housing. A set of multispectral cameras and a set of lidar are respectively installed at the top of the support plate. A set of gears is installed at each end of the rotating shaft. Two sets of optical bars are provided between the UAV body and the base plate. The sliding rod is slidably connected to the base plate. A rack is provided at the top of the sliding rod, and the rack meshes with the gears. A lifting plate is provided at the top of the rack, and the lifting plate is slidably connected to the two sets of optical bars. A limiting plate is provided on the sliding rod. The system is slidably connected to two sets of optical bars, each with a set of springs mounted on its outer side, all positioned above the lifting plate. When not in use, the drone lands on the ground, and its weight raises the sliding rod, which in turn raises the rack. The rack meshes with the gear, causing the support plate to rotate 180 degrees, allowing the multispectral camera and LiDAR to enter the housing. When the propellers rotate, the drone takes off, and the lifting plate, under the spring force, moves the rack downwards, causing the gear to rotate 180 degrees, further moving the multispectral camera and LiDAR below the support plate for field scanning. The drone's takeoff and landing facilitate the storage and relocation of the multispectral camera and LiDAR, improving convenience.

[0009] Preferably, the locking assembly includes a fixing block, a socket, a fixing seat, and a wing bolt. The fixing block is installed at the bottom of the support plate, and the fixing block is provided with a socket. The fixing seat is installed at the bottom of the housing, and the fixing seat is threadedly connected to the wing bolt. When not in use, the operator rotates the wing bolt, and the wing bolt, with the cooperation of the fixing seat, inserts into the socket on the fixing block, thereby locking the support plate and improving convenience.

[0010] Preferably, it also includes a pointed cone portion, and the bottom end of the slide bar is provided with a pointed cone portion; when the main body of the drone lands on the ground, the support plate makes contact with the ground to improve anti-slip performance.

[0011] Preferably, the assembly also includes a positioning rod, a sealing plate, bolts, and a handle. The housing has an inspection port and a positioning rod. The sealing plate has a positioning hole. The sealing plate seals the inspection port with bolts. A handle is installed on the sealing plate. Workers can inspect the interior through the inspection port. After inspection, the positioning rod is inserted into the positioning hole to complete the initial installation of the sealing plate. The bolts then secure the sealing plate. Workers can move the sealing plate using the handle, improving convenience.

[0012] Preferably, it also includes an observation window, which is installed on the sealing plate; staff can observe the internal condition of the box through the observation window, improving convenience.

[0013] Preferably, it further includes an annular flat washer, and a set of annular flat washers are respectively sleeved on each group of bolts; the bolts increase the contact area with the sealing plate through the annular flat washers, improving the connection firmness.

[0014] A method for monitoring wheat harvesting loss includes the following steps: Step 1: Data collection and preprocessing Piezoelectric sensors capture grain impact signals, image sensors identify missed cut wheat ears and damaged grains, and environmental sensors synchronously collect parameters such as header height and traveling speed; Wavelet transform is used to remove high-frequency noise, and the grain signal and straw signal are separated through adaptive filtering; Step 2: Real-time calculation of loss rate Determine single-grain or multi-grain impact based on the pulse interval time of piezoelectric signals: If the pulse interval time t < ams, it is determined as single grain; If the total duration T > bms, accumulate multi-grain loss according to N = N + ⌊T / b⌋; Combined with the missed cut rate and breakage rate identified by image recognition, comprehensively calculate the total loss rate; Step 3: Dynamic regulation When the loss rate exceeds the threshold (such as > 1.2%), the control module automatically adjusts parameters: Increase the rotational speed of the threshing cylinder and reduce the threshing gap; Lower the header height to 5 - 15 cm, and optimize the speed ratio of the reel (α = 1.1 ∼ 1.2 × v traveling); Adjust the opening of the cleaning sieve and the air volume of the fan to balance the impurity content rate and cleaning loss; Step 4: Calibration and verification Calibrate the sensor sensitivity under a simulated vibration environment; During field operation, regularly compare with the results of manual sampling to calibrate the monitoring accuracy.

[0015] Compared with existing technologies, the advantages of this invention are as follows: The UAV monitoring module scans the field, identifies lodged areas and high-density areas, and generates a "loss risk heat map." The loss risk heat map is imported into the harvester navigation system, prioritizing the harvesting of high-risk areas; a piezoelectric grain sensor is installed at the tail of the cleaning screen, using vibration isolation devices (such as rubber pads) to reduce mechanical vibration interference, and detects grain impact signals through piezoelectric elements; an image sensor is deployed at the header and threshing outlet, using deep learning algorithms to identify missed wheat ears and damaged grains; a loss rate calculation unit calculates the loss rate of each sampling area according to a formula for wheat harvesting; a parameter optimization unit is used to establish a mapping model between the loss rate and operating parameters (such as threshing gap, screen opening, and fan airflow), and automatically adjusts the harvester parameters through a PID controller; a human-machine interface terminal displays real-time loss rate, operating parameter suggestions, and alarm information, supports data export and analysis, and realizes a high-precision, real-time, and automatically adjustable monitoring system. Attached Figure Description

[0016] Figure 1 This is a system block diagram of the present invention; Figure 2 This is a first isometric schematic diagram of the UAV monitoring unit of the present invention; Figure 3 This is a second isometric schematic diagram of the UAV monitoring unit of the present invention; Figure 4 This is an exploded structural diagram of the box body and the main body of the drone; Figure 5 This is an enlarged structural diagram of the drone's main body and base plate; Figure 6 This is an enlarged structural diagram of the lifting plate and gears, etc. Figure 7 This is an enlarged structural diagram of the support plate and observation window, among other structures. Figure 8 yes Figure 7 A partially enlarged structural diagram of section A in the middle; Figure 9 This is an exploded structural diagram of the box body and support plates, etc. Figure 10 It is an enlarged structural diagram of a multispectral camera and springs, etc.

[0017] The attached diagram shows the following markings: 101, UAV body; 102, propeller; 103, support rod; 104, base plate; 105, housing; 201, pivot; 202, multispectral camera; 203, lidar; 204, gear; 205, optical bar; 206, slide bar; 207, lifting plate; 208, rack; 209, spring; 210, limiting plate; 211, support plate; 212, pointed cone; 301, fixing block; 302, insertion hole; 303, fixing seat; 304, wing bolt; 401, positioning rod; 402, sealing plate; 403, bolt; 404, handle; 405, observation window; 406, annular flat gasket. Detailed Implementation

[0018] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. The present invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0019] Example 1 like Figures 1 to 10 The present invention discloses a wheat harvest loss monitoring system, comprising an unmanned aerial vehicle (UAV) monitoring module, a multimodal sensor array, a signal processing module, an intelligent control module, and a human-machine interaction terminal. The drone monitoring module is used to scan fields, identify lodged areas and high-density areas, and generate a "loss risk heat map". The loss risk heat map is imported into the harvester navigation system to prioritize the harvesting of high-risk areas. The multimodal sensor array includes a piezoelectric grain sensor and an image sensor; Piezoelectric grain sensor: Installed at the tail of the cleaning screen, it uses vibration isolation devices (such as rubber pads) to reduce mechanical vibration interference and detects grain impact signals through piezoelectric elements; Image sensors: deployed at the header and threshing exit, using deep learning algorithms to identify missed ears of wheat and damaged grains; The intelligent control module includes a loss rate calculation unit and a parameter optimization unit; The loss rate calculation unit calculates the loss rate for each sampling area for rice, wheat, and corn grain harvesting according to the following formula:

[0020] In the formula: S—loss rate, in terms of %; W—grain loss of rice (corn, wheat) in the sampling area, in terms of grams; M—grain yield of rice (corn, wheat) per unit area, in terms of kilograms per mu; L is taken as 2 meters or the working width of a commonly used combine harvester in the local area, in terms of meters; The parameter optimization unit is used to establish a mapping model between the loss rate and operating parameters (such as threshing gap, sieve opening, and fan air volume), and automatically adjusts the harvester parameters through a PID controller; The human-machine interface terminal is used to display real-time loss rate, suggested operating parameters and alarm information, and supports data export and analysis; The multimodal sensor array also includes a header height sensor, a reel speed sensor, and a travel speed sensor, which work together to collect operating parameters in real time. The drone monitoring module includes a drone body 101, propellers 102, support rods 103, a base plate 104, and a housing 105. The drone body 101 is equipped with multiple sets of propellers 102, and two sets of support rods 103 are provided at the bottom of the drone body 101. Each set of support rods 103 is provided with a base plate 104 at its bottom. The housing 105 is installed at the bottom of the drone body 101 and also includes a crop lodging monitoring component. The crop lodging monitoring component is installed inside the housing 105. The crop lodging monitoring component includes a rotating shaft 201, a multispectral camera 202, a lidar 203, gears 204, an optical bar 205, a sliding rod 206, a lifting plate 207, a rack 208, a spring 209, a limiting plate 210, and a support plate 211. The support plate 211 is rotatably mounted on the bottom inner side of the housing 105 with the cooperation of the rotating shaft 201. The support plate 211 seals the bottom of the housing 105. A set of multispectral cameras 202 and a set of lidar 203 are respectively installed at the top of the support plate 211. A set of gears 204 are installed at each end of the rotating shaft 201. 4. Two sets of optical bars 205 are provided between the main body 101 of the drone and the base plate 104. A sliding rod 206 is slidably connected to the base plate 104. A rack 208 is provided at the top of the sliding rod 206. The rack 208 meshes with the gear 204. A lifting plate 207 is provided at the top of the rack 208. The lifting plate 207 is slidably connected to the two sets of optical bars 205. A limiting plate 210 is provided on the sliding rod 206. The limiting plate 210 is slidably connected to the two sets of optical bars 205. A set of springs 209 is respectively fitted on the outside of each set of optical bars 205. The springs 209 are all above the lifting plate 207. It also includes a locking assembly, which is provided between the housing 105 and the support plate 211; The locking assembly includes a fixing block 301, a socket 302, a fixing seat 303, and a wing bolt 304. The fixing block 301 is installed at the bottom of the support plate 211, and the socket 302 is provided on the fixing block 301. The fixing seat 303 is installed at the bottom of the housing 105, and the fixing seat 303 is threadedly connected to the wing bolt 304. It also includes a pointed cone portion 212, and the bottom end of the slide rod 206 is provided with a pointed cone portion 212; It also includes a positioning rod 401, a sealing plate 402, a bolt 403 and a handle 404. The box 105 is provided with an inspection port, the box 105 is provided with a positioning rod 401, the sealing plate 402 is provided with a positioning hole, the sealing plate 402 seals the inspection port by the bolt 403, and the handle 404 is installed on the sealing plate 402. It also includes an observation window 405, which is installed on the sealing plate 402; It also includes annular flat washers 406, with a set of annular flat washers 406 fitted on each set of bolts 403.

[0021] In this embodiment, when not in use, the drone body 101 lands on the ground. The weight of the drone body 101 causes the slide bar 206 to rise, thereby raising the rack 208. Since the rack 208 meshes with the gear 204, the support plate 211 rotates 180 degrees, allowing the multispectral camera 202 and the lidar 203 to enter the housing 105. When the multiple propellers 102 rotate, the drone body 101 takes off. The lifting plate 207, under the action of the spring 209, causes the rack 208 to move downward, thereby causing the gear 204 to rotate 180 degrees, further moving the multispectral camera 202 and the lidar 203 below the support plate 211, so that the multispectral camera 202 and the lidar 203 can perform field scanning. The multispectral camera 202 and lidar 203 are stored and moved around by the take-off and landing of the main body of the drone 101. The multispectral camera 202 and lidar 203 identify lodged areas and high-density areas and generate a "loss risk heat map". The loss risk heat map is imported into the harvester navigation system to prioritize the harvesting of high-risk areas. The piezoelectric grain sensor is installed at the tail of the cleaning screen. It uses a vibration isolation device (such as a rubber pad) to reduce mechanical vibration interference and detects the grain impact signal through the piezoelectric element. Image sensors are deployed at the header and threshing exit to identify missed ears of wheat and damaged grains based on deep learning algorithms; The loss rate calculation unit calculates the loss rate of each sampling area for wheat harvesting according to the formula. The parameter optimization unit is used to establish a mapping model between the loss rate and operating parameters (such as threshing gap, sieve opening, and fan air volume), and automatically adjusts the harvester parameters through the PID controller. The human-machine interaction terminal displays the real-time loss rate, operating parameter suggestions, and alarm information, and supports data export and analysis, realizing a high-precision, real-time, and automatically adjustable monitoring system.

[0022] Example 2 The present invention provides a method for monitoring wheat harvest loss, comprising the following steps: Step 1: Data acquisition and preprocessing. The piezoelectric sensor captures the grain impact signal, the image sensor identifies the missed cut wheat ears and damaged grains, and the environmental sensor synchronously collects parameters such as the header height and traveling speed; Wavelet transform is used to remove high-frequency noise, and the grain signal and the stalk signal are separated by adaptive filtering; Step 2: Real-time calculation of the loss rate Determine single or multiple grain impacts based on the pulse interval time of the piezoelectric signal: If the pulse interval time t < a ms, it is determined as single grain; If the total duration T > b ms, the multiple grain loss is accumulated as N = N + ⌊T / b⌋; Combined with the missed cut rate and breakage rate identified by image recognition, the total loss rate is comprehensively calculated; Step 3: Dynamic regulation When the loss rate exceeds the threshold (such as > 1.2%), the control module automatically adjusts the parameters: Increase the threshing cylinder speed and reduce the threshing gap; Lower the header height to 5 - 15 cm and optimize the reel speed ratio (α = 1.1 ∼ 1.2 × v traveling); Adjust the cleaning sieve opening and the fan air volume to balance the impurity content rate and the cleaning loss; Step 4: Calibration and verification Calibrate the sensor sensitivity under a simulated vibration environment; During field operation, regularly compare with the results of manual sampling to calibrate the monitoring accuracy.

[0023] The main functions realized by this invention are: 1. Multimodal fusion monitoring: Combining piezoelectric sensing and image recognition to take into account the full-dimensional detection of grain loss, missed cut, and breakage; 2. Adaptive signal processing: The dynamic threshold algorithm and wavelet filtering improve the signal accuracy in complex environments; 3. Intelligent closed-loop control: Construct a regulation model based on policy standards and historical data to achieve automatic optimization of minimizing the loss rate; 4. The real-time monitoring accuracy is improved to over 95%, far exceeding manual sampling (error ±20%); 5. Through automatic regulation, the loss rate can be reduced to below 1.0%; 6. Reduce the dependence on the operator's experience and adapt to different wheat varieties and field conditions.

[0024] The wheat harvest loss monitoring system and method of the present invention can be installed, connected or set up in a common mechanical manner, and can be implemented as long as it can achieve its beneficial effect. The multispectral camera 202 and lidar 203 of the wheat harvest loss monitoring system and method of the present invention are commercially available. Technical personnel in the industry only need to install and operate them according to the accompanying instruction manual, without requiring any creative work from those skilled in the art.

[0025] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A wheat harvest loss monitoring system, comprising a drone monitoring module, a multimodal sensor array, a signal processing module, an intelligent control module, and a human-machine interaction terminal; characterized in that, The drone monitoring module is used to scan fields, identify lodged areas and high-density areas, and generate a "loss risk heat map". The loss risk heat map is imported into the harvester navigation system to prioritize the harvesting of high-risk areas. The multimodal sensor array includes a piezoelectric grain sensor and an image sensor; Piezoelectric grain sensor: Installed at the tail of the cleaning screen, it uses a vibration isolation device to reduce mechanical vibration interference and detects grain impact signals through piezoelectric elements; Image sensors: deployed at the header and threshing exit, using deep learning algorithms to identify missed ears of wheat and damaged grains; The intelligent control module includes a loss rate calculation unit and a parameter optimization unit; The loss rate calculation unit calculates the loss rate for each sampling area for rice, wheat, and corn grain harvesting according to the following formula: In the formula: S—loss rate, in terms of %; W—grain loss of rice (corn, wheat) in the sampling area, in terms of grams; M—grain yield of rice (corn, wheat) per unit area, in terms of kilograms per mu; L is taken as 2 meters or the working width of a commonly used combine harvester in the local area, in terms of meters; The parameter optimization unit is used to establish a mapping model between the loss rate and the operating parameters, and automatically adjusts the harvester parameters through the PID controller; The human-machine interface terminal is used to display real-time loss rate, suggested operating parameters, and alarm information, and supports data export and analysis.

2. The wheat harvest loss monitoring system as described in claim 1, characterized in that, The multimodal sensor array also includes a header height sensor, a reel speed sensor, and a travel speed sensor, which work together to collect operational parameters in real time.

3. The wheat harvest loss monitoring system as described in claim 1, characterized in that, The drone monitoring module includes a drone body (101), propellers (102), support rods (103), base plate (104), and housing (105). The drone body (101) is equipped with multiple sets of propellers (102). The drone body (101) has two sets of support rods (103) at its bottom. Each set of support rods (103) has a base plate (104) at its bottom. The drone body (101) has a housing (105) at its bottom. The module also includes a crop lodging monitoring component. The crop lodging monitoring component is installed inside the housing (105).

4. The wheat harvest loss monitoring system as described in claim 3, characterized in that, The crop lodging monitoring component includes a rotating shaft (201), a multispectral camera (202), a lidar (203), a gear (204), a light bar (205), a sliding rod (206), a lifting plate (207), a rack (208), a spring (209), a limiting plate (210) and a support plate (211). The support plate (211) is rotatably installed at the inner bottom of the box body (105) in cooperation with the rotating shaft (201). The support plate (211) seals the bottom end of the box body (105). A group of multispectral cameras (202) and a group of lidars (203) are respectively arranged at the top end of the support plate (211). A group of gears (204) are respectively installed at both ends of the rotating shaft (201). Two groups of light bars (205) are arranged between the drone main body (101) and the bottom plate (104). The sliding rod (206) is slidably connected to the bottom plate (104). A rack (208) is arranged at the top end of the sliding rod (206). The rack (208) meshes with the gear (204). A lifting plate (207) is arranged at the top end of the rack (208). The lifting plate (207) is slidably connected to the two groups of light bars (205). A limiting plate (210) is arranged on the sliding rod (206). The limiting plate (210) is slidably connected to the two groups of light bars (205). A group of springs (209) are respectively sleeved on the outer sides of each group of light bars (205). The springs (209) are all above the lifting plate (207).

5. A wheat harvest loss monitoring system as described in claim 4, characterized in that, The locking component includes a fixed block (301), a jack (302), a fixed seat (303) and a wing bolt (304). The fixed block (301) is installed at the bottom end of the support plate (211). The jack (302) is arranged on the fixed block (301). The fixed seat (303) is installed at the bottom end of the box body (105). The fixed seat (303) is threadedly connected to the wing bolt (304).

6. The wheat harvest loss monitoring system as described in claim 4, characterized in that, It further includes a tapered portion (212). The tapered portion (212) is arranged at the bottom end of the sliding rod (206).

7. A wheat harvest loss monitoring system as described in claim 5, characterized in that, It further includes a positioning rod (401), a sealing plate (402), bolts (403) and a handle (404). An inspection opening is arranged on the box body (105). The positioning rod (401) is arranged on the box body (105). A positioning hole is arranged on the sealing plate (402). The sealing plate (402) seals the inspection opening through the bolts (403). The handle (404) is installed on the sealing plate (402).

8. A wheat harvest loss monitoring system as described in claim 7, characterized in that, It further includes an observation window (405). The observation window (405) is installed on the sealing plate (402).

9. A wheat harvest loss monitoring system as described in claim 7, characterized in that, It further includes an annular flat washer (406). A group of annular flat washers (406) are respectively sleeved on each group of bolts (403).

10. A method for monitoring wheat harvest loss, characterized in that, It includes the following steps, Step 1: Data acquisition and preprocessing The piezoelectric sensor captures the grain impact signal. The image sensor identifies the missed cut wheat ears and damaged grains. The environmental sensor synchronously collects parameters such as the height of the cutting table and the driving speed; Wavelet transform is used to remove high-frequency noise, and the grain signal and the stalk signal are separated through adaptive filtering; Step 2: Real-time calculation of the loss rate Determine single-grain or multi-grain impact based on the pulse interval time of the piezoelectric signal: If the pulse interval time t < ams, it is determined as single grain; If the total duration T > bms, accumulate the multi-particle loss according to N = N + ⌊T / b⌋. The total loss rate is calculated by combining the missed cutting rate and the breakage rate of image recognition. Step 3: Dynamic Regulation When the loss rate exceeds a threshold (e.g., >1.2%), the control module automatically adjusts the parameters: Increase the speed of the threshing drum and reduce the threshing gap; Lower the header height to 5-15cm and optimize the reel speed ratio (α=1.1~1.2×v travel). Adjust the opening of the cleaning screen and the air volume of the fan to balance the impurity content and cleaning loss; Step 4: Calibration and Verification Calibrate sensor sensitivity under simulated vibration conditions; During field operations, the results are regularly compared with those obtained from manual sampling to calibrate the monitoring accuracy.

Images