Soil deep scarification farming agricultural equipment based on artificial intelligence

Abstract

The invention relates to the technical field of soil deep scarification farming agricultural equipment, and discloses artificial intelligence-based soil deep scarification farming agricultural equipment which comprises a machine head, a connecting rod is connected to the rear part of the machine head through a connecting device, a cross rod is fixedly connected to one end, away from the machine head, of the connecting rod, and a soil turning shovel head is arranged at the bottom of the cross rod; the rear portion of the machine head is connected to the output end of a hydraulic cylinder through a fixing device, the side, away from the output end, of the hydraulic cylinder is connected to the upper surface of a connecting rod through a fixing device, and an environment and terrain monitoring module is arranged on the front portion of the machine head. A data transmission module, a decision support module, a supervision module and a central processing unit are arranged in the machine head. According to the invention, the environment and terrain monitoring module can monitor the terrain and the state under the soil in real time during working, so that the problem that the traditional soil subsoiling farming agricultural machine is damaged due to the fact that the distribution condition of stones under the surface of the soil layer cannot be observed frequently is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of soil deep tillage agricultural machinery and equipment, and in particular to a soil deep tillage agricultural machinery and equipment based on artificial intelligence. Background Art

[0002] Deep tillage machinery is used for deep soil cultivation. By breaking up the plow base and loosening the soil structure, it significantly improves soil fertility and the growing environment for crops. This type of equipment is typically equipped with a high-strength plow or deep tiller, capable of operating at depths exceeding 30 centimeters. This effectively reduces soil compaction, enhances air permeability, and improves water and fertilizer retention. It is particularly suitable for farmland where the tillage layer has become shallow due to long-term use of shallow tillage machinery. Key models include deep tillers, rotary tillers, and rotary tillers. Deep tillers reduce damage to surface vegetation by loosening the soil in intervals. Rotary tillers thoroughly turn the soil to eliminate weeds, pests, and diseases. Rotary tillers combine soil pulverization with fertilizer mixing. Deep tillage machinery has become crucial for increasing grain production and achieving sustainable agricultural development, particularly in applications such as protecting the black soil in Northeast China, conserving moisture in dryland areas of North China, and renovating hilly terraced fields in southern China.

[0003] Traditional deep soil tillage machinery often relies on manual adjustment, but the soil conditions are mostly obscured, and operators cannot observe the distribution of stones below the soil surface, which can cause damage to the tillage machinery. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides an artificial intelligence-based deep soil loosening and tillage agricultural machinery equipment, which solves the problem that traditional deep soil loosening and tillage agricultural machinery often relies on manual adjustment, but the situation in the soil is mostly obscured, and the operator cannot observe the distribution of stones below the soil surface, which will cause damage to the tillage agricultural machinery.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an artificial intelligence-based soil deep tillage agricultural machinery equipment, including a head, the rear of the head is connected to a connecting rod through a connecting device, the end of the connecting rod away from the head is fixedly connected to a cross bar, and a tillage shovel head is provided at the bottom of the cross bar, the rear of the head is connected to the output end of the hydraulic cylinder through a fixing device, and the side of the hydraulic cylinder away from the output end is connected to the upper surface of the connecting rod through a fixing device, the front of the head is provided with an environment and terrain monitoring module, the interior of the head is provided with a data transmission module, a decision support module, a supervision module and a central processing unit, a hydraulic adjustment system is provided inside the hydraulic cylinder, and a soil parameter sensor module is provided inside the tillage shovel head, the environment and terrain monitoring module cooperates with the soil parameter sensor module to collect land and environmental conditions in real time and then transmits them to the decision support module, the supervision module and the central processing unit through the data transmission module for processing and decision-making, and controls the hydraulic adjustment system to adjust the working state of the hydraulic cylinder after processing and decision-making.

[0006] Preferably, the connecting device includes a mounting block, the outer wall of the mounting block is fixedly connected to the rear of the machine head, the interior of the mounting block is fixedly connected with a rotating shaft, a mounting hole is opened on the side of the connecting rod close to the machine head, and the rotating shaft is located inside the mounting hole.

[0007] Preferably, the fixing device includes a first connecting ring, a second connecting ring, a third connecting ring and a fourth connecting ring, the machine head is fixedly connected to the rear of the machine head, the output end of the hydraulic cylinder is fixedly connected to the second connecting ring, the first connecting ring is connected to the second connecting ring, the third connecting ring is provided on the side of the hydraulic cylinder away from the output end, the fourth connecting ring is provided on the top surface of the connecting rod, and the third connecting ring is connected to the fourth connecting ring.

[0008] Preferably, the environment and terrain monitoring module includes a ground-penetrating radar unit, a multispectral imaging sensor unit, a radar terrain monitoring unit, and a binocular vision plus line laser monitoring unit. The ground-penetrating radar unit is used to locate the position of stones in the range 0-50 cm below the soil surface and identify the size of stones. The multispectral imaging sensor unit is used to capture the reflection information of the soil and objects therein in multiple spectral bands. The radar terrain monitoring unit is used to perform three-dimensional modeling of the terrain. The binocular vision plus line laser monitoring unit is used to generate a three-dimensional point cloud of the soil structure.

[0009] Preferably, the soil parameter sensor module includes a near-infrared spectroscopy soil composition sensor unit, a frequency domain reflectometry moisture sensor unit, a temperature sensor unit and a soil bulk density detection unit. The near-infrared spectroscopy soil composition sensor unit is used to detect the nitrogen, phosphorus and potassium content in the soil, the frequency domain reflectometry moisture sensor unit is used to detect the volume, mass and moisture content of the soil, the temperature sensor unit is used to detect the temperature conditions at different depths in the soil, and the soil bulk density detection unit is used to detect the deep loosening condition of the plow bottom layer.

[0010] Preferably, the data transmission module includes a WiFi unit and a 5G communication unit, the WiFi unit is used for near-field data transmission, and the 5G communication unit is used for remote data interaction.

[0011] Preferably, the decision support module includes a machine learning algorithm unit and a predictive maintenance recommendation unit, wherein the machine learning algorithm unit is used to analyze soil degradation, fertility changes and pollution trends, and the predictive maintenance recommendation unit is used to monitor the status of key components in real time to predict potential failure risks.

[0012] Preferably, the supervision module includes a real-time monitoring unit for the operation trajectory, an intelligent calculation unit for the operation area, and a quality assessment unit. The real-time monitoring unit for the operation trajectory is used to record the trajectory of the deep plowing operation, the intelligent calculation unit for the operation area is used to calculate the deep plowing operation area, and the quality assessment unit is used to inspect and evaluate the quality of the deep plowing completion section.

[0013] Preferably, the machine learning algorithm unit adopts a linear regression algorithm when analyzing the soil degradation trend, and its formula is: Y = β0 + β1X1 + β2X2 + ... + β n X n +∈where Y represents the predicted value of soil degradation degree; X1,X2,…,X n They represent the characteristic variables related to soil degradation obtained from the environment and terrain monitoring module and the soil parameter sensor module, β0 is the intercept, β1, β2,…, β n are the regression coefficients of the corresponding characteristic variables, which are obtained by training a large amount of historical soil degradation data; ∈ is the error term, which is used to represent the difference between the model prediction value and the actual value.

[0014] Preferably, the machine learning algorithm unit uses the principal component analysis algorithm when analyzing soil fertility changes, and the formula for calculating the principal component is: Among them, z i is the i-th principal component; X j is the original soil fertility related variable; w ijIt is the load of the i-th principal component on the j-th original variable, obtained by performing eigendecomposition and other operations on the covariance matrix of soil fertility data, and is used to determine the contribution of each original variable to the principal component.

[0015] Working principle: The machine head drives the connecting rod to pull the tilling shovel head to move, thereby deeply loosening the soil. The ground penetrating radar unit in the environment and terrain monitoring module inside the machine head locates the position of 0-50cm stones under the soil and identifies the size. The multispectral imaging sensor unit captures the multispectral reflection information of the soil and objects inside the soil to identify internal debris. The radar terrain monitoring unit models the terrain in three dimensions. The binocular vision plus line laser monitoring unit generates a three-dimensional point cloud of the soil structure. The collected data is then transmitted to the central processing unit. The central processing unit sends instructions to the hydraulic adjustment system based on the comprehensive judgment results. The hydraulic adjustment system adjusts the working state of the hydraulic cylinder according to the instructions. By changing the extension and contraction of the hydraulic cylinder, the connecting rod is driven to change the angle with the rotating shaft as the center, and then the operating depth and angle of the tilling shovel head are adjusted to avoid collision with stones.

[0016] At the same time, the soil parameter sensor module's near-infrared spectroscopy soil composition sensor unit detects soil nitrogen, phosphorus, and potassium content; the frequency domain reflectometry moisture sensor unit detects soil volume, mass, and moisture content; the temperature sensor unit detects soil temperature at different depths; and the soil bulk density detection unit detects the deep loosening of the plow bottom layer. The central processor performs preliminary processing and analysis on the collected land and environmental data, and the machine learning algorithm unit of the decision support module analyzes soil degradation, fertility changes, and pollution trends.

[0017] The predictive maintenance recommendation unit monitors the status of key components and predicts potential failure risks. The supervisory module's real-time operation trajectory monitoring unit records operation trajectories, the intelligent operation area calculation unit calculates the operation area, and the quality assessment unit evaluates operation quality. The processing results of each module are fed back to the central processor, which makes a comprehensive assessment to ensure that the agricultural machinery is always in optimal operating condition.

[0018] The present invention provides an artificial intelligence-based deep soil tillage agricultural machinery. It has the following beneficial effects:

[0019] 1. In the present invention, the environment and terrain monitoring module inside the machine head can monitor the terrain and the status of the soil during operation in real time, and then can adjust the working status of the hydraulic cylinder in real time to adjust the depth of the tillage shovel head to avoid stones in the soil, thereby solving the problem that traditional soil deep loosening and tillage agricultural machinery often relies on manual adjustment, but the situation in the soil is mostly obscured, and the operator cannot observe the distribution of stones below the soil surface, which will cause damage to the tillage agricultural machinery.

[0020] 2. In the present invention, the machine learning algorithm unit in the decision support module accurately analyzes soil degradation, fertility changes and pollution trends based on a large amount of data, providing data support for the formulation of scientific soil improvement, fertilization and pollution prevention and control strategies, and realizing the sustainable utilization of soil resources.

[0021] 3. In the present invention, the data transmission module includes a WiFi unit and a 5G communication unit, which are used for near-field and remote data interaction respectively, to ensure the stable transmission of various types of data and meet the needs of different operating scenarios. With the help of 5G communication, agricultural experts or producers can remotely obtain data and accurately control agricultural machinery, realize remote management, improve agricultural production management efficiency, and promote the development of smart agriculture. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of the three-dimensional structure of the present invention;

[0023] Figure 2 This is a schematic diagram of the local structure of the mounting block of the present invention;

[0024] Figure 3 For the present invention Figure 1 A schematic diagram of the structure at center A;

[0025] Figure 4 For the present invention Figure 1 A magnified schematic diagram of the structure at B in the middle;

[0026] Figure 5 A schematic diagram of the system module structure of the present invention;

[0027] Figure 6 Schematic diagram of the system structure of the environment and terrain monitoring module of the present invention;

[0028] Figure 7 FIG is a schematic diagram of the system structure of the soil parameter sensor module of the present invention;

[0029] Figure 8 This is a schematic diagram of the system structure of the data transmission module of the present invention;

[0030] Figure 9 Schematic diagram of the system structure of the decision support module of the present invention;

[0031] Figure 10 Schematic diagram of the system structure of the supervision module of the present invention.

[0032] Among them, 1. Machine head; 2. Hydraulic cylinder; 3. Mounting block; 4. Connecting rod; 5. Cross bar; 6. Rotating shaft; 7. Mounting hole; 8. First connecting ring; 9. Second connecting ring; 10. Third connecting ring; 11. Fourth connecting ring; 12. Soil-turning shovel head. DETAILED DESCRIPTION

[0033] The following will clearly and completely describe the technical solution of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0034] Please see the attached Figure 1 -Attached Figure 5 The embodiment of the present invention provides an artificial intelligence-based soil deep tillage agricultural machinery equipment, including a head 1, the rear of the head 1 is connected to a connecting rod 4 through a connecting device, the end of the connecting rod 4 away from the head 1 is fixedly connected to a cross bar 5, and a tillage shovel head 12 is provided at the bottom of the cross bar 5, the rear of the head 1 is connected to the output end of the hydraulic cylinder 2 through a fixing device, and the side of the hydraulic cylinder 2 away from the output end is connected to the upper surface of the connecting rod 4 through a fixing device, the front of the head 1 is provided with an environment and terrain monitoring module, the interior of the head 1 is provided with a data transmission module, a decision support module, a supervision module and a central processing unit, the interior of the hydraulic cylinder 2 is provided with a hydraulic adjustment system, the interior of the tillage shovel head 12 is provided with a soil parameter sensor module, the environment and terrain monitoring module cooperates with the soil parameter sensor module to collect land and environmental conditions in real time and then transmits them to the decision support module, the supervision module and the central processing unit through the data transmission module for processing and decision-making, and controls the hydraulic adjustment system to adjust the working state of the hydraulic cylinder 2 after processing and decision-making.

[0035] Specifically, the machine head 1 can provide power and place related modules; the connecting device can be used to install the connecting rod 4 at the rear of the machine head 1; the connecting rod 4 can be used to connect the cross bar 5; the cross bar 5 can be used to install the tillage shovel head 12; the tillage shovel head 12 can be used to deep loosen the soil; the fixing device can be used to fix the hydraulic cylinder 2 between the rear of the machine head 1 and the top surface of the connecting rod 4; the environment and terrain monitoring module can locate the position and size of 0-50cm stones in the soil, obtain multi-spectral reflectance information of soil and objects, three-dimensionally model the terrain and generate a three-dimensional point cloud of soil structure, providing data support for agricultural machinery operations; the data transmission module can realize near-field data transmission and remote data interaction to ensure the transmission of land and environmental collection data, processing decision results and equipment status information; the decision support module can analyze soil degradation, fertility changes, pollution trends and monitor the status of key components to predict potential failure risks, providing scientific and reasonable decision-making basis for agricultural machinery operations and equipment maintenance; the supervision module can monitor the operation trajectory The system monitors and records deep tillage trajectories in real time, intelligently calculates the operating area, and inspects and evaluates the operating quality of the work by a quality assessment unit, ensuring that agricultural machinery operations are standardized, efficient, and up to standard. The central processor receives data collected by the environment and terrain monitoring module and the soil parameter sensor module. After processing and analysis, the collaborative decision support module and the supervision module make a comprehensive judgment on the data, and then controls the hydraulic adjustment system to adjust the working state of the hydraulic cylinder 2, thereby realizing intelligent control of the operation of agricultural machinery. The hydraulic adjustment system can adjust the working state of the hydraulic cylinder 2 according to the instructions of the central processor, and then adjust the operating depth and angle of the tillage shovel head 12 to adapt to different soil conditions and tillage requirements, thereby solving the problem that traditional soil deep tillage agricultural machinery often relies on manual adjustment, but the soil conditions are mostly obscured, and the operator cannot observe the distribution of stones below the soil surface, which will cause damage to the tillage agricultural machinery. The soil parameter sensor module can detect the nitrogen, phosphorus, and potassium content, volume, mass, moisture content, temperature at different depths, and deep tillage conditions of the plow bottom layer in the soil, providing soil data support for agricultural machinery operation decisions.

[0036] Please see the attached Figure 1 -Attached Figure 2 The connecting device includes a mounting block 3, the outer wall of the mounting block 3 is fixedly connected to the rear of the machine head 1, the interior of the mounting block 3 is fixedly connected with a rotating shaft 6, a mounting hole 7 is opened on the side of the connecting rod 4 close to the machine head 1, and the rotating shaft 6 is located inside the mounting hole 7.

[0037] Specifically, the mounting block 3 can connect the connecting rod 4 and the rear part of the machine head 1; the connecting rod 4 can be installed through the mounting hole 7 through the rotating shaft 6, so that the connecting rod 4 can rotate around the rotating shaft 6, thereby achieving the purpose of changing the depth of the soil-turning shovel head 12.

[0038] Please see the attached Figure 1 -Attached Figure 4 The fixing device includes a first connecting ring 8, a second connecting ring 9, a third connecting ring 10 and a fourth connecting ring 11. The machine head 1 is fixedly connected to the rear part of the machine head 1. The output end of the hydraulic cylinder 2 is fixedly connected with the second connecting ring 9. The first connecting ring 8 is connected to the second connecting ring 9. The third connecting ring 10 is provided on the side of the hydraulic cylinder 2 away from the output end. The fourth connecting ring 11 is provided on the top surface of the connecting rod 4. The third connecting ring 10 is connected to the fourth connecting ring 11.

[0039] Specifically, by connecting the first connecting ring 8 with the second connecting ring 9, the output end of the hydraulic cylinder 2 can be fixed to the rear of the machine head 1; by connecting the third connecting ring 10 with the fourth connecting ring 11, the side of the hydraulic cylinder 2 away from the output end can be fixed to the top surface of the connecting rod 4; thereby, the angle of the connecting rod 4 can be driven by the extension and contraction of the hydraulic cylinder 2, thereby changing the working depth of the earth-turning shovel head 12.

[0040] Please see the attached Figure 6 The environment and terrain monitoring module includes a ground-penetrating radar unit, a multispectral imaging sensor unit, a radar terrain monitoring unit, and a binocular vision plus line laser monitoring unit. The ground-penetrating radar unit is used to locate the position of stones 0-50 cm below the soil surface and identify the size of stones. The multispectral imaging sensor unit is used to capture the reflection information of the soil and objects in it under multiple spectral bands. The radar terrain monitoring unit is used to make three-dimensional modeling of the terrain. The binocular vision plus line laser monitoring unit is used to generate three-dimensional point clouds of soil structure.

[0041] Specifically, the ground-penetrating radar unit is used to locate the position of 0-50cm stones under the soil and identify their size, providing data basis for agricultural machinery to avoid stones in advance and prevent damage during operation, thereby ensuring the safety and stability of agricultural machinery operations; the multispectral imaging sensor unit is used to capture the reflection information of the soil and objects therein in multiple spectral bands, which can be used to identify foreign matter in the soil, analyze soil composition and fertility conditions, and provide data basis for agricultural machinery to intelligently adjust its operation strategy; the radar terrain monitoring unit is used to perform three-dimensional modeling of the terrain, allowing agricultural machinery to automatically adjust the operating height and angle according to the terrain undulations, ensuring that the tillage shovel head 12 always maintains the appropriate tillage depth, improving operation accuracy and efficiency, and reducing operation deviations caused by terrain changes; the binocular vision plus line laser monitoring unit can generate a three-dimensional point cloud of soil structure, providing data for analyzing soil structure conditions and judging the degree of soil looseness, assisting agricultural machinery in accurately adjusting tillage methods to optimize the deep loosening effect of the soil.

[0042] Please see the attached Figure 7The soil parameter sensor module includes a near-infrared spectroscopy soil composition sensor unit, a frequency domain reflectometry moisture sensor unit, a temperature sensor unit and a soil bulk density detection unit. The near-infrared spectroscopy soil composition sensor unit is used to detect the nitrogen, phosphorus and potassium content in the soil. The frequency domain reflectometry moisture sensor unit is used to detect the volume, mass and moisture content of the soil. The temperature sensor unit is used to detect the temperature conditions at different depths in the soil. The soil bulk density detection unit is used to detect the deep loosening condition of the plow bottom layer.

[0043] Specifically, the near-infrared spectroscopy soil composition sensor unit is used to detect the nitrogen, phosphorus and potassium content in the soil, providing data for the agricultural machinery decision-making system to accurately adjust the fertilization strategy, achieve scientific fertilization, and improve soil fertility and crop yields; the frequency domain reflectometry moisture sensor unit is used to detect soil volume, quality and moisture content, and can accurately obtain relevant data. In terms of soil volume detection, it can provide accurate basic data for land planning and irrigation volume calculation, and can monitor its dynamic changes in real time, and can draw spatial distribution maps to analyze regional differences; for soil quality, fertility and other conditions can be judged based on the test results to provide a basis for improvement measures; in terms of moisture content detection, it can provide key information for precise irrigation, reasonably control irrigation volume, and improve water resource utilization efficiency, ultimately improving agricultural production decision-making optimization, and improving agricultural production efficiency and sustainability; the temperature sensor unit is used to detect the temperature at different depths in the soil, and can accurately obtain temperature data at each depth, which can help agricultural practitioners understand the vertical distribution of soil temperature and provide a basis for crop planting planning, such as judging Determine the depth suitable for sowing or transplanting; help analyze the changing patterns of soil temperature in different seasons and periods, so as to take timely measures to deal with the impact of extreme temperatures on crop roots, such as pre-heating the soil during low temperatures in winter; can also provide data support for the study of microbial activity in the soil, because microbial activity is affected by soil temperature, thereby optimizing the soil ecological environment and ensuring the healthy growth of crops; the soil bulk density detection unit is used to detect the deep loosening of the plow bottom layer, which can accurately obtain soil bulk density data. According to the soil bulk density data, it can be intuitively determined whether the plow bottom layer is effectively loosened. When the deep loosening effect is good, the soil bulk density decreases, which means that the plow bottom layer structure is broken and the soil pores increase, thereby improving the soil's air permeability and water permeability, which is conducive to the downward extension of crop roots. At the same time, the detection unit can provide agricultural producers with key information such as deep loosening depth and range, so as to analyze the differences in deep loosening effects of different plots, and then optimize the deep loosening operation plan, ensure the quality of deep loosening, improve the soil's water and fertilizer retention capacity, create a high-quality soil environment for crop growth, and promote high and stable crop yields.

[0044] Please see the attached Figure 8 The data transmission module includes a WiFi unit and a 5G communication unit. The WiFi unit is used for near-field data transmission, and the 5G communication unit is used for remote data interaction.

[0045] Specifically, using WiFi units for near-field data transmission enables efficient and stable data exchange between devices. In agricultural scenarios, real-time data collected by soil sensors, weather stations, and other equipment can be quickly transmitted to terminal devices or data processing centers. This allows farmers or agricultural technicians to obtain timely information such as soil moisture, temperature, bulk density, and meteorological conditions, allowing them to make accurate decisions based on this data, such as rationally arranging irrigation, fertilization, and farming operation times. This improves the intelligence and refinement of agricultural production and improves agricultural production efficiency and benefits. Using 5G communication units for remote data interaction enables high-speed, low-latency data transmission. A large amount of soil parameters, meteorological data, crop growth status, and other information collected by various on-site sensors can be transmitted to a remote data processing platform or user terminal. This enables agricultural experts or producers, even in remote locations, to obtain comprehensive and accurate data in real time, allowing them to conduct timely and precise remote control of agricultural production, such as remotely adjusting irrigation system flow, starting and stopping fertilization equipment, and optimizing greenhouse environmental parameters. This greatly improves the efficiency of remote management of agricultural production and effectively promotes the efficient development of smart agriculture.

[0046] Please see the attached Figure 9 The decision support module includes a machine learning algorithm unit and a predictive maintenance recommendation unit. The machine learning algorithm unit is used to analyze soil degradation, fertility changes and pollution trends. The predictive maintenance recommendation unit is used to monitor the status of key components in real time to predict potential failure risks.

[0047] Specifically, the machine learning algorithm unit is used to analyze soil degradation, fertility changes and pollution trends. Based on a large amount of historical and real-time monitoring data, it can accurately identify soil degradation patterns, predict changes in soil fertility in different time periods in the future, and effectively judge the development trend of soil pollution. By establishing a model, the algorithm can mine the complex correlations between data, identify the key factors affecting soil conditions, and provide data support for the formulation of targeted soil improvement measures, scientific fertilization plans and pollution prevention and control strategies; the predictive maintenance recommendation unit monitors the status of key components in real time and predicts potential failure risks. It can continuously collect component operating parameters and identify anomalies in advance through algorithm analysis, thereby providing early warning at the budding stage of failure, enabling the maintenance team to plan maintenance tasks in advance, prepare spare parts, and carry out maintenance operations accurately. This can significantly reduce unexpected equipment downtime, reduce maintenance costs, and extend the service life of equipment.

[0048] Please see the attached Figure 10 The supervision module includes a real-time monitoring unit for operation trajectory, an intelligent calculation unit for operation area, and a quality assessment unit. The real-time monitoring unit for operation trajectory is used to record the trajectory of deep plowing operation, the intelligent calculation unit for operation area is used to calculate the deep plowing operation area, and the quality assessment unit is used to inspect and evaluate the quality of the completed deep plowing section.

[0049] Specifically, by recording the deep plowing operation trajectory through the real-time monitoring unit of the operation trajectory, the actual coverage of the deep plowing operation can be reflected, and it can be clarified which areas have completed deep plowing and which areas still need to be operated, so as to avoid missed or repeated operations. By analyzing the trajectory data, the uniformity of the operation can be evaluated, and it can be judged whether the deep plowing depth and spacing meet the standards, which is helpful to adjust the operation parameters in time. At the same time, these records provide a basis for subsequent farmland management, facilitate the traceability of the operation, and provide strong data support for optimizing the deep plowing operation plan and improving the operation quality and efficiency. The operation area intelligent calculation unit is used to calculate the deep plowing operation area, and can obtain the actual area data of the operation area in real time and accurately. By comparing with the preset operation range, it can quickly determine whether the operation is fully covered and promptly discover the omission or out-of-range operation. This provides an accurate basis for the billing of agricultural machinery operations and avoids Unclear area accounting will lead to economic disputes. At the same time, it will help agricultural production managers to accurately grasp the progress of deep tillage, reasonably arrange subsequent farming activities, evaluate operation efficiency based on area data, provide key support for optimizing agricultural machinery scheduling and operation planning, and improve the level of refinement of agricultural production management. The quality assessment unit is used to inspect and evaluate the quality of the completed deep tillage section. According to established quality standards, it can conduct quantitative analysis on key indicators such as deep tillage depth, soil looseness, and plot flatness. By comparing field measurements with standard values, it can accurately judge whether the operation meets the requirements, identify non-compliant areas and specific problems, which provides a basis for timely corrective measures and ensures that the quality of deep tillage meets agricultural production needs. At the same time, the evaluation data will help summarize experience, provide support for the optimization of subsequent operation plans, promote the continuous improvement of deep tillage operation quality, and lay a good soil foundation for crop growth.

[0050] The machine learning algorithm unit uses a linear regression algorithm to analyze soil degradation trends, and its formula is: Y = β0 + β1X1 + β2X2 + … + β n X n +∈where Y represents the predicted value of soil degradation degree; X1,X2,…,X n They represent the characteristic variables related to soil degradation obtained from the environment and terrain monitoring module and the soil parameter sensor module, β0 is the intercept, β1, β2,…, β n are the regression coefficients of the corresponding characteristic variables, which are obtained by training a large amount of historical soil degradation data; ∈ is the error term, which is used to represent the difference between the model prediction value and the actual value.

[0051] Specifically, with the help of linear regression algorithm formula, the multi-category characteristic variables obtained from the environment and terrain monitoring module and the soil parameter sensor module can be comprehensively considered to give a quantitative prediction value of the degree of soil degradation and provide data support for soil management decision-making; through the regression coefficients β1, β2,…, β n, which can evaluate the direction and magnitude of the impact of each characteristic variable on the degree of soil degradation, help to identify key influencing factors, and provide a basis for formulating targeted soil protection measures; the regression coefficient is obtained based on a large amount of historical soil degradation data training, which ensures the objectivity and reliability of the model, makes the soil degradation analysis based on data-driven, and improves the scientific nature of decision-making; the error term ∈ can measure the difference between the model predicted value and the actual value, which helps to evaluate the accuracy and reliability of the model and provide a reference for model optimization and improvement.

[0052] When analyzing soil fertility changes, the machine learning algorithm unit uses the principal component analysis algorithm. The formula for calculating the principal component is: Among them, z i is the i-th principal component; X j is the original soil fertility related variable; w ij It is the load of the i-th principal component on the j-th original variable, obtained by performing eigendecomposition and other operations on the covariance matrix of soil fertility data, and is used to determine the contribution of each original variable to the principal component.

[0053] Specifically, multiple original soil fertility-related variables are converted into a few principal components, which reduces the data dimension and analysis complexity while retaining most of the information; key information is extracted from a large number of original variables to make the main characteristics of soil fertility changes more prominent, which is convenient for analysis and understanding; the contribution of each original variable to the principal component is determined through the principal component loading, which helps to identify the key factors affecting soil fertility changes; the principal components after dimensionality reduction can be used for subsequent modeling analysis, reducing the input variables of the model and improving the efficiency and stability of the model.

[0054] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An artificial intelligence-based soil deep tillage agricultural machinery, characterized in that: It includes a machine head, the rear part of the machine head is connected to a connecting rod through a connecting device, the end of the connecting rod away from the machine head is fixedly connected to a cross bar, the bottom of the cross bar is provided with a tiller head, the rear part of the machine head is connected to the output end of the hydraulic cylinder through a fixing device, the side of the hydraulic cylinder away from the output end is connected to the upper surface of the connecting rod through a fixing device, the front part of the machine head is provided with an environment and terrain monitoring module, the interior of the machine head is provided with a data transmission module, a decision support module, a supervision module and a central processing unit, the interior of the hydraulic cylinder is provided with a hydraulic adjustment system, the interior of the tiller head is provided with a soil parameter sensor module, the environment and terrain monitoring module cooperates with the soil parameter sensor module to collect land and environmental conditions in real time and then transmits them to the decision support module, the supervision module and the central processing unit through the data transmission module for processing and decision-making, and controls the hydraulic adjustment system to adjust the working state of the hydraulic cylinder after processing and decision-making.

2. The artificial intelligence-based soil subsoiling agricultural machinery according to claim 1, characterized in that: The connecting device includes a mounting block, the outer wall of the mounting block is fixedly connected to the rear of the machine head, the interior of the mounting block is fixedly connected to a rotating shaft, a mounting hole is opened on the side of the connecting rod close to the machine head, and the rotating shaft is located inside the mounting hole.

3. The artificial intelligence-based soil subsoiling agricultural machinery according to claim 1, characterized in that: The fixing device includes a first connecting ring, a second connecting ring, a third connecting ring and a fourth connecting ring. The machine head is fixedly connected to the rear part of the machine head. The output end of the hydraulic cylinder is fixedly connected to the second connecting ring. The first connecting ring is connected to the second connecting ring. The third connecting ring is provided on the side of the hydraulic cylinder away from the output end. The fourth connecting ring is provided on the top surface of the connecting rod. The third connecting ring is connected to the fourth connecting ring.

4. The artificial intelligence-based soil subsoiling agricultural machinery according to claim 3, characterized in that: The environment and terrain monitoring module includes a ground-penetrating radar unit, a multispectral imaging sensor unit, a radar terrain monitoring unit, and a binocular vision plus line laser monitoring unit. The ground-penetrating radar unit is used to locate the position of stones 0-50 cm below the soil surface and identify the size of the stones. The multispectral imaging sensor unit is used to capture the reflection information of the soil and objects therein in multiple spectral bands. The radar terrain monitoring unit is used to perform three-dimensional modeling of the terrain. The binocular vision plus line laser monitoring unit is used to generate a three-dimensional point cloud of the soil structure.

5. The artificial intelligence-based soil subsoiling agricultural machinery equipment according to claim 1, characterized in that: The soil parameter sensor module includes a near-infrared spectroscopy soil composition sensor unit, a frequency domain reflectometry moisture sensor unit, a temperature sensor unit, and a soil bulk density detection unit. The near-infrared spectroscopy soil composition sensor unit is used to detect the nitrogen, phosphorus, and potassium content in the soil. The frequency domain reflectometry moisture sensor unit is used to detect the volume, mass, and moisture content of the soil. The temperature sensor unit is used to detect the temperature conditions at different depths in the soil. The soil bulk density detection unit is used to detect the deep loosening conditions of the plow bottom layer.

6. The artificial intelligence-based soil subsoiling agricultural machinery equipment according to claim 5, characterized in that: The data transmission module includes a WiFi unit and a 5G communication unit, the WiFi unit is used for near-field data transmission, and the 5G communication unit is used for remote data interaction.

7. The artificial intelligence-based soil subsoiling agricultural machinery equipment according to claim 6, characterized in that: The decision support module includes a machine learning algorithm unit and a predictive maintenance recommendation unit. The machine learning algorithm unit is used to analyze soil degradation, fertility changes and pollution trends. The predictive maintenance recommendation unit is used to monitor the status of key components in real time and predict potential failure risks.

8. The artificial intelligence-based soil subsoiling agricultural machinery equipment according to claim 7, characterized in that: The supervision module includes a real-time monitoring unit for operation trajectory, an intelligent calculation unit for operation area, and a quality assessment unit. The real-time monitoring unit for operation trajectory is used to record the trajectory of deep plowing operation, the intelligent calculation unit for operation area is used to calculate the deep plowing operation area, and the quality assessment unit is used to inspect and evaluate the quality of the deep plowing completion section.

9. The artificial intelligence-based soil subsoiling agricultural machinery according to claim 7, characterized in that: The machine learning algorithm unit uses a linear regression algorithm when analyzing soil degradation trends, and its formula is: Y = β0 + β1X1 + β2X2 + ... + β n X n +∈where Y represents the predicted value of soil degradation degree; X1,X2,…,X n They represent the characteristic variables related to soil degradation obtained from the environment and terrain monitoring module and the soil parameter sensor module, β0 is the intercept, β1, β2,…, β n are the regression coefficients of the corresponding characteristic variables, which are obtained by training a large amount of historical soil degradation data; ∈ is the error term, which is used to represent the difference between the model prediction value and the actual value.

10. The artificial intelligence-based soil subsoiling agricultural machinery equipment according to claim 7, characterized in that: When analyzing soil fertility changes, the machine learning algorithm unit uses the principal component analysis algorithm to calculate the formula for the principal component: Among them, z i is the i-th principal component; X j is the original soil fertility related variable; w ij It is the load of the i-th principal component on the j-th original variable, obtained by performing eigendecomposition and other operations on the covariance matrix of soil fertility data, and is used to determine the contribution of each original variable to the principal component.

Images