An intelligent control strategy for cooperative operation of an electro-hydraulic suspension device for high-horsepower tractors

Through the intelligent control strategy of collaborative operation of electro-hydraulic suspension devices of high-horsepower tractors, data acquisition and machine learning algorithms are used to adjust the suspension device parameters in real time, solving the problem that traditional control systems are difficult to cope with changes in the operating environment, and improving the operating efficiency and automation level.

CN118761036BActive Publication Date: 2024-12-06NANJING AGRI MECHANIZATION INST MIN OF AGRI
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411245885.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-06
Publication Date
2024-12-06
Estimated Expiration
2044-09-06

AI Technical Summary

Technical Problem

The traditional tractor suspension control system lacks flexibility and adaptability, and it is difficult to deal with changes in the operating environment in real time, resulting in difficulty in ensuring operational efficiency and quality.

Method used

The coordinated operation intelligent control strategy of the electro-hydraulic suspension device of the high-horsepower tractor is adopted. Through data acquisition, preprocessing, machine learning algorithms and feedback control, the suspension height, hydraulic pressure and operation speed are adjusted in real time, and a dynamic control strategy is generated to adapt to different operating conditions.

Benefits of technology

Significantly improve the accuracy and efficiency of operations, reduce dependence on operators, reduce errors and resource waste, improve the automation and intelligence level of agricultural machinery, optimize resource use, and ensure that the tractor remains the most in various environments. Excellent performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118761036B_ABST
    Figure CN118761036B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of agricultural machinery, and specifically to a collaborative operation intelligent control strategy for an electro-hydraulic suspension device of a high-horsepower tractor, comprising the following steps: S1: establishing an operation task library; S2: real-time collection of suspension device information during tractor operation; S3: preprocessing the collected real-time data; S4: analyzing the data set, combining the parameters in the operation task library, and generating a preliminary suspension device control strategy; S5: data interaction and collaborative control with other operation units; S6: identifying the deviation between the operating state and the expected control parameters; S7: dynamically adjusting the preliminary control strategy. The present invention, by implementing sensing technology and machine learning algorithms, realizes real-time and precise adjustment of the operating parameters of the tractor suspension device, significantly improves the level of operation automation and intelligence, and thus effectively improves the efficiency and economic benefits of agricultural operations.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of agricultural machinery, and in particular to a cooperative operation intelligent control strategy for an electro-hydraulic suspension device of a high-horsepower tractor. Background Art

[0002] In modern agricultural operations, the efficiency and accuracy of high-horsepower tractors and their attachments have a direct impact on operating costs and output quality. In particular, the control strategy of the electro-hydraulic suspension device is crucial to achieving precision agriculture and improving operating efficiency. Traditional tractor suspension system control relies more on the operator's experience and manual adjustment, which not only increases the complexity of operation, but also makes it difficult to adapt to the changing operating environment and soil conditions. For example, different operating tasks such as ploughing, sowing and harvesting have different requirements for suspension height, hydraulic pressure and operating speed, which require frequent and precise adjustments to adapt to different operating conditions.

[0003] Although there are some automated control solutions in the prior art, these solutions often fail to fully consider the actual impact of environmental changes on operations, resulting in difficulty in ensuring operating efficiency and quality. In addition, existing control systems often lack sufficient flexibility and adaptability and cannot respond to changes in operating conditions in real time, such as uneven terrain and fluctuations in soil moisture. These factors may lead to differences in operating results. Therefore, the tractor suspension control strategy in the prior art still has a lot of room for improvement in automation and intelligence. Summary of the invention

[0004] Based on the above objectives, the present invention provides an intelligent control strategy for the coordinated operation of an electro-hydraulic suspension device of a high-horsepower tractor.

[0005] A collaborative operation intelligent control strategy for an electro-hydraulic suspension device of a high-horsepower tractor comprises the following steps:

[0006] S1: Establish an operation task library and define corresponding suspension device control strategy parameters for different types of agricultural operation tasks;

[0007] S2: Through the data acquisition module, the position information of the suspension device, the pressure information of the hydraulic system, the operating speed, the terrain data and the environmental information during the operation of the tractor are collected in real time and stored in the data storage unit;

[0008] S3: The data processing module pre-processes the real-time data collected in the data storage unit, including removing noise and abnormal data, and performing feature extraction and data fusion to generate a unified data set;

[0009] S4: The intelligent decision-making module based on the machine learning algorithm analyzes the data set and generates a preliminary suspension control strategy based on the parameters in the task library, including specific control parameters of suspension height, hydraulic pressure and operation speed;

[0010] S5: The collaborative operation module receives the preliminary control strategy generated by the intelligent decision-making module, and performs data exchange and collaborative control with other operation units to ensure coordination and consistency among the operation units;

[0011] S6: The feedback control module receives the operating state data collected in real time from the data acquisition module, processes and analyzes the data, and identifies the deviation between the operating state and the expected control parameters;

[0012] S7: The intelligent decision-making module dynamically adjusts the preliminary control strategy based on the deviations identified by the feedback control module to cope with environmental changes and operating condition changes that occur during the actual operation process.

[0013] Optionally, the S1 specifically includes:

[0014] S11: Classify different types of agricultural tasks, including tillage, sowing, and harvesting tasks;

[0015] S12: For each operation task, collect relevant operation parameters and historical data, including information on terrain characteristics, soil type, operation depth, operation speed, and operation tools, and organize and standardize the data to form a database;

[0016] S13: Based on the collected historical data and relevant operating parameters, the optimal suspension control strategy parameters for various operating tasks are determined using statistical analysis and machine learning algorithm analysis, including specific parameters of suspension height, hydraulic pressure and operating speed;

[0017] S14: The control strategy parameters obtained through analysis are stored in the operation task library, and a corresponding relationship between the parameters and specific operation tasks is established for use in the actual operation process.

[0018] Optionally, the S2 specifically includes:

[0019] S21: The data acquisition module is equipped with a variety of sensors, including a position sensor, a pressure sensor, a speed sensor, a terrain scanner and an environmental monitor; wherein the position sensor is used to monitor the position of the suspension device in real time; the pressure sensor is used to monitor the working pressure of the hydraulic system; the speed sensor is used to record the operating speed of the tractor; the terrain scanner is used to obtain the terrain data of the current operating area; and the environmental monitor is used to collect environmental information such as temperature and humidity;

[0020] S22: The data collected by each sensor is transmitted to the central processing unit in real time through the data collection unit, and the central processing unit is used to perform preliminary time stamp marking and classification processing on the received data;

[0021] S23: During the data collection process, implement data filtering algorithms to remove noise or abnormal data, the filtering algorithms include signal smoothing, outlier detection and data normalization;

[0022] S24: storing the preliminarily processed and verified data in a data storage unit;

[0023] S25: Establish data index. Each data item is marked with a unique index and a label of the relevant job task, so that the data can be quickly associated with the scheduled job task.

[0024] Optionally, the S3 specifically includes:

[0025] S31: extracting real-time data from the data storage unit, including position information of the suspension device, pressure information of the hydraulic system, operating speed, terrain data and environmental information;

[0026] S32: applying a data cleaning algorithm to remove noise and abnormal data from the extracted data, specifically using a median filter algorithm and a mean filter algorithm to smooth the data and eliminate instantaneous noise; and using a statistical analysis method to detect and remove abnormal values;

[0027] S33: feature extraction, by analyzing the cleaned data, extracting features, including the displacement change rate of the suspension device; the fluctuation range of the hydraulic system pressure; the instantaneous speed and acceleration of the tractor; the terrain slope and surface roughness; and the change trend of the ambient temperature and humidity;

[0028] S34: performing data fusion, specifically using a Kalman filter algorithm to fuse multi-sensor data, and performing dimensionality reduction processing on multi-dimensional feature data to reduce redundant data, so as to form a comprehensive feature data set;

[0029] S35: Generate a unified data set, store the fused comprehensive feature data set in a data storage unit, and add a timestamp and a task label to each data set.

[0030] Optionally, the S4 specifically includes:

[0031] S41: The intelligent decision module receives the preprocessed and fused data set;

[0032] S42: Matching the received data set with the preset parameters in the job task library to generate a data set matching the current job conditions;

[0033] S43: Analyze the matched data set based on the random forest machine learning algorithm to determine the suspension control parameters that best match the current operating conditions;

[0034] S44: Based on the analysis results of the random forest algorithm, a preliminary suspension control strategy is generated, which specifies the specific parameters of suspension height, hydraulic pressure and operating speed to best adapt to the current operating environment.

[0035] Optionally, the S42 specifically includes:

[0036] S421: Preliminarily matching the preprocessed and fused data set with the preset parameters in the job task library;

[0037] S422: Calculate the difference between each parameter in the current data set and the preset parameters in the job task library, the formula is: ,in, Represents the difference between the i-th parameter in the current data set and the j-th preset parameter in the job task library. Represents the i-th parameter value in the current data set, Indicates the jth preset parameter value in the job task library;

[0038] S423: Calculate the total difference of each task to determine the most matching task. The following formula is: ,in, represents the total difference of the jth job task in the job task library, and n represents the total number of parameters;

[0039] S424: Select the job task with the smallest total difference as the most matching job task at present, the formula is: ,in, Indicates the most matching job task index, that is, the total difference Minimal homework tasks, is the minimum function;

[0040] S425: Using the preset parameters of the job task that best matches the data set as preliminary control parameters.

[0041] Optionally, the S43 specifically includes:

[0042] S431: constructing a training data set, including historical operation data and corresponding control parameters, the data set including suspension height, hydraulic pressure, operation speed, corresponding environmental conditions and operation results;

[0043] S432: Use the training data set to train a random forest model. The random forest consists of multiple decision trees, and each decision tree is constructed using the following formula: , in, represents the decision tree model, X represents the input feature vector, represents the i-th decision tree, represents the corresponding weight, and n is the number of decision trees;

[0044] S433: Input the matched data set in S42 into the trained random forest model for prediction. The random forest model outputs the optimal control parameters by integrating the prediction results of all decision trees. The specific formula is: ,in, represents the predicted optimal control parameter vector, N is the total number of decision trees, Represents the prediction result of the jth decision tree;

[0045] S434: The random forest model generates control parameters of the suspension device according to the input data set, including suspension height, hydraulic pressure and operating speed.

[0046] Optionally, the S5 specifically includes:

[0047] S51: The collaborative operation module receives the preliminary control strategy generated by the intelligent decision module, which includes the suspension height , hydraulic pressure and operating speed Specific parameters of

[0048] S52: transmitting the received preliminary control strategy to each operation unit, including the tillage unit and the sowing unit, through the internal bus;

[0049] S53: Each operating unit receives the preliminary control strategy and adjusts its own operating state according to the strategy parameters;

[0050] S54: The collaborative operation module monitors the operation status and feedback data of each operation unit in real time through sensors, including actual tillage depth and actual sowing speed;

[0051] S55: The collaborative operation module aggregates the feedback data of each operation unit and compares it with the preliminary control strategy, and adjusts the preliminary control strategy through the built-in collaborative optimization algorithm to ensure coordination and consistency among the operation units;

[0052] S56: According to the adjusted control strategy, the operating parameters of each operating unit are dynamically adjusted to ensure the coordination and consistency of each unit in the collaborative operation.

[0053] Optionally, the S6 specifically includes:

[0054] S61: The feedback control module receives the real-time collected operation status data from the data acquisition module, including the suspension height , hydraulic system pressure Information and operation speed information;

[0055] S62: The feedback control module performs preliminary processing on the received data, and the preliminary processing includes data filtering, denoising and data calibration;

[0056] S63: The feedback control module compares the actual operation state data after preliminary processing with the expected parameters in the preliminary control strategy generated by the intelligent decision module to identify the deviation between the operation state and the expected control parameters;

[0057] Calculate the deviation of the suspension height , the formula is: ,in, The expected suspension height generated by the intelligent decision module, is the actual suspension height;

[0058] Calculate the deviation of hydraulic pressure , the formula is: ,in, The expected hydraulic pressure generated by the intelligent decision module, is the actual hydraulic pressure;

[0059] Calculate the deviation of the working speed , the formula is: ,in, The expected operation speed generated for the intelligent decision-making module, The actual operating speed.

[0060] Optionally, the S7 specifically includes:

[0061] S71: The intelligent decision module receives the deviation data calculated by the feedback control module, including the suspension height deviation, the hydraulic pressure deviation and the operating speed deviation;

[0062] S72: The intelligent decision-making module analyzes the deviation data to determine the trend and potential causes of the deviation;

[0063] S73: Based on the analysis results, the intelligent decision module uses an adaptive adjustment algorithm to dynamically adjust the preliminary control strategy. Specifically, when the deviation is a persistent deviation, the intelligent decision module will adjust the corresponding control parameters to reduce the deviation; when the deviation is a temporary deviation, the intelligent decision module will monitor and record the deviation data and will not adjust the parameters immediately; the dynamic adjustment specifically uses the proportional integral differential control algorithm to correct the parameters:

[0064] The formula for updating the suspension height control parameters is: ;

[0065] The formula for updating the hydraulic pressure control parameters is: ;

[0066] The formula for updating the operation speed control parameter is: , in, , and are the proportional, integral and differential coefficients of the PID controller respectively;

[0067] S74: The adjusted control parameters are transmitted to the collaborative operation module and related operation units to ensure that all operation units operate according to the new control strategy.

[0068] Beneficial effects of the present invention:

[0069] The present invention, through the implementation of advanced sensing technology and machine learning algorithms, can accurately monitor and adjust the operating parameters of the suspension device, such as suspension height, hydraulic pressure and operating speed, in real time to ensure that these parameters are always adapted to the current operating conditions. This fine control can significantly improve the accuracy and efficiency of operations, reduce dependence on operators, reduce errors and waste of resources caused by improper operation, and significantly improve the automation and intelligence level of agricultural machinery operations.

[0070] The present invention can dynamically adjust the control strategy according to the real-time data collected during the operation process through the feedback regulation mechanism to cope with the changes in the operating environment. This not only increases the adaptability of the operating process, but also further optimizes the use of resources, such as saving fuel and time. Through the comprehensive application of these technologies, it can ensure that the tractor can maintain optimal performance in various operating environments, thereby improving the economic benefits and sustainability of the entire agricultural production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0072] Figure 1 A schematic diagram of an intelligent control strategy for collaborative operations according to an embodiment of the present invention;

[0073] Figure 2 A schematic diagram of a process for dynamically adjusting a control strategy according to an embodiment of the present invention;

[0074] Figure 3 Schematic diagram of sensor installation position according to an embodiment of the present invention. DETAILED DESCRIPTION

[0075] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0076] It should be noted that the references to "one embodiment", "embodiment", "exemplary embodiments", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0077] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0078] like Figure 1-Figure 3 As shown, a collaborative operation intelligent control strategy for an electro-hydraulic suspension device of a high-horsepower tractor includes the following steps:

[0079] S1: Establish an operation task library to define corresponding suspension device control strategy parameters for different types of agricultural operation tasks (such as ploughing, sowing, harvesting, etc.), including suspension height, hydraulic pressure, and operation speed;

[0080] S2: Through the data acquisition module, the position information of the suspension device, the pressure information of the hydraulic system, the operating speed, the terrain data and the environmental information during the operation of the tractor are collected in real time and stored in the data storage unit;

[0081] S3: The data processing module pre-processes the real-time data collected in the data storage unit, including removing noise and abnormal data, and performing feature extraction and data fusion to generate a unified data set;

[0082] S4: The intelligent decision-making module based on the machine learning algorithm analyzes the data set and generates a preliminary suspension control strategy based on the parameters in the task library, including specific control parameters of suspension height, hydraulic pressure and operation speed;

[0083] S5: The collaborative operation module receives the preliminary control strategy generated by the intelligent decision-making module, and performs data exchange and collaborative control with other operation units (such as tillage units and sowing units) to ensure coordination and consistency among the operation units and improve overall operation efficiency;

[0084] S6: The feedback control module receives the operating state data collected in real time from the data acquisition module, processes and analyzes the data, and identifies the deviation between the operating state and the expected control parameters;

[0085] S7: The intelligent decision-making module dynamically adjusts the preliminary control strategy based on the deviations identified by the feedback control module to cope with the environmental changes and operating conditions that may occur during the actual operation process, ensuring the optimal performance and stable operation of the suspension device.

[0086] The establishment of the job task library in S1 specifically includes:

[0087] S11: Classify different types of agricultural tasks, including tillage, sowing, and harvesting tasks. Each type of task has specific operational requirements and environmental characteristics.

[0088] S12: For each operation task, collect relevant operation parameters and historical data, including information on terrain characteristics, soil type, operation depth, operation speed, and operation tools, and organize and standardize the data to form a database for analysis;

[0089] S13: Based on the collected historical data and relevant operating parameters, statistical analysis and machine learning algorithm analysis are used to determine the optimal suspension control strategy parameters for various operating tasks, including specific parameters of suspension height, hydraulic pressure and operating speed; among them, the suspension height is used to determine the optimal height of the suspension device according to the needs of different operating tasks to ensure the best contact between the operating tool and the ground; the hydraulic pressure is determined by combining the soil type and terrain characteristics to determine the optimal pressure setting of the hydraulic system to provide sufficient support and buffering effect; the operating speed is used to determine the optimal operating speed of the tractor according to the nature of the operating task and the ground conditions to achieve efficient operation and reduce energy consumption;

[0090] S14: The control strategy parameters obtained through analysis are stored in an operation task library, and a corresponding relationship between the parameters and specific operation tasks is established for calling during the actual operation process; the operation task library established through the above specific steps enables a high-horsepower tractor to adjust the operating parameters of the suspension device in real time according to historical data and optimal control strategy parameters when performing different types of agricultural operation tasks, which not only improves the operation efficiency and quality, but also reduces energy consumption and mechanical wear, and enhances the adaptability and stability of the system.

[0091] S2 specifically includes:

[0092] S21: The data acquisition module is equipped with a variety of sensors, including position sensors, pressure sensors, speed sensors, terrain scanners and environmental monitors; wherein the position sensor is used to monitor the position of the suspension device in real time; the pressure sensor is used to monitor the working pressure of the hydraulic system; the speed sensor is used to record the operating speed of the tractor; the terrain scanner is used to obtain the terrain data of the current operating area; the environmental monitor is used to collect environmental information such as temperature and humidity;

[0093] S22: The data collected by each sensor is transmitted to the central processing unit in real time through the data collection unit, and the central processing unit is used to perform preliminary time stamp marking and classification processing on the received data;

[0094] S23: During the data collection process, data filtering algorithms are implemented to remove possible noise or abnormal data to ensure the accuracy and reliability of the data. The filtering algorithms include signal smoothing, outlier detection, and data normalization;

[0095] S24: The data that has been preliminarily processed and verified is stored in a data storage unit. The data storage unit uses a high-efficiency database that supports high-speed data writing and query operations and has data backup and recovery functions to ensure the security and integrity of the data during the operation process;

[0096] S25: Establish data indexes to optimize data retrieval efficiency and facilitate future data analysis. Each data item is marked with a unique index and a label of the relevant work task, so that the data can be quickly associated with the predetermined work task. Through the above steps, the data acquisition module can efficiently and accurately collect and store key data during the tractor operation process, including the position information of the suspension device, the pressure information of the hydraulic system, the operating speed, and the terrain and environmental information. The real-time collection and storage of this data not only provides accurate input for the intelligent decision-making module, but also optimizes the execution of the work task, improves the work efficiency and reduces potential operating errors, thereby enhancing the overall performance and reliability of the system.

[0097] Preprocessing in S3 specifically includes:

[0098] S31: extracting real-time data from the data storage unit, including position information of the suspension device, pressure information of the hydraulic system, operating speed, terrain data and environmental information;

[0099] S32: Apply data cleaning algorithms to remove noise and abnormal data from the extracted data. Specifically, use median filtering algorithms and mean filtering algorithms to smooth the data and eliminate instantaneous noise. Statistical analysis methods, such as the Z-Score method, are used to detect and remove abnormal values ​​to ensure the accuracy and consistency of the data.

[0100] S33: feature extraction, by analyzing the cleaned data, extracting features, including the displacement change rate of the suspension device; the fluctuation range of the hydraulic system pressure; the instantaneous speed and acceleration of the tractor; the terrain slope and surface roughness; and the change trend of the ambient temperature and humidity;

[0101] S34: performing data fusion, specifically using a Kalman filter algorithm to fuse multi-sensor data, and performing dimensionality reduction processing on multi-dimensional feature data to reduce redundant data, so as to form a comprehensive feature data set;

[0102] S35: Generate a unified data set, store the fused comprehensive feature data set in the data storage unit, and add a timestamp and task label to each data set to ensure the traceability and relevance of the data; and perform data verification and consistency check on the generated unified data set to ensure the integrity and accuracy of the data set, and prepare it for use by the intelligent decision-making module; through the above specific steps, the real-time data in the data storage unit is preprocessed, including removing noise and abnormal data, and performing feature extraction and data fusion to generate a unified data set. This process not only improves the quality and reliability of the data, but also enhances the efficiency of data analysis and processing, and provides high-quality input data for the intelligent decision-making module, thereby significantly improving the accuracy and operating efficiency of the suspension control strategy, and ensuring the optimal performance and stable operation of the system under various operating conditions.

[0103] S4 specifically includes:

[0104] S41: The intelligent decision module receives the preprocessed and fused data set, which contains key features such as the displacement change rate of the suspension device, the pressure fluctuation range of the hydraulic system, and the speed change of the tractor;

[0105] S42: Matching the received data set with the preset parameters in the operation task library to generate a data set matching the current operation conditions, which is used for prediction in the intelligent decision-making module. The operation task library predefines the control parameters required for various agricultural operation tasks (such as ploughing, sowing, and harvesting), including suspension height, hydraulic pressure, and operation speed;

[0106] S43: A machine learning algorithm based on random forest analyzes the matched data set to determine the suspension control parameters that best match the current operating conditions. Random forest improves the accuracy and robustness of predictions by building multiple decision trees and combining their results.

[0107] S44: Based on the analysis results of the random forest algorithm, a preliminary suspension control strategy is generated, which specifies the specific parameters of suspension height, hydraulic pressure and operating speed to best adapt to the current operating environment; through the above steps, the intelligent decision-making module can automatically generate a suspension control strategy that adapts to the current operating conditions based on accurate and comprehensive data analysis and combined with the preset parameters in the operating task library. This method uses advanced machine learning technology to ensure the high accuracy of the control strategy and the optimization of operation, significantly improves operating efficiency and safety, reduces energy consumption and mechanical wear, and enhances the overall performance and reliability of the system.

[0108] S42 specifically includes:

[0109] S421: Preliminarily matching the preprocessed and fused data set with preset parameters in the operation task library, where the operation task library contains ideal parameter ranges of suspension height, hydraulic pressure, and operation speed for each operation task;

[0110] S422: Calculate the difference between each parameter in the current data set and the preset parameters in the job task library, the formula is: ,in, Represents the difference between the i-th parameter in the current data set and the j-th preset parameter in the job task library. Represents the i-th parameter value in the current data set, Indicates the jth preset parameter value in the job task library;

[0111] S423: Calculate the total difference of each task to determine the most matching task. The following formula is: ,in, represents the total difference of the jth job task in the job task library, and n represents the total number of parameters;

[0112] S424: Select the job task with the smallest total difference as the most matching job task at present, the formula is: ,in, Indicates the most matching job task index, that is, the total difference Minimal homework tasks, is the minimum function;

[0113] S425: The preset parameters of the work task that best matches the data set are used as preliminary control parameters, including suspension height, hydraulic pressure and work speed, and are used for the next step of intelligent decision-making analysis; through the above steps, the current data set can be accurately matched with the preset parameters in the work task library, so as to select the control parameters that best suit the current working conditions, and the accuracy and reliability of the matching process are ensured by calculating the difference between each parameter and comprehensively calculating the total difference, which not only improves the efficiency of generating the control strategy, but also ensures the accuracy and stability of the operation, thereby further improving the operating performance and overall reliability of the tractor.

[0114] S43 specifically includes:

[0115] S431: Construct a training data set, including historical operation data and corresponding control parameters. The data set includes suspension height , hydraulic pressure , Operation speed and the corresponding environmental conditions and operating results;

[0116] S432: Use the training data set to train a random forest model. The random forest consists of multiple decision trees, and each decision tree is constructed using the following formula: , in, represents the decision tree model, X represents the input feature vector (including parameters such as H, P, V, etc.), represents the i-th decision tree, represents the corresponding weight, and n is the number of decision trees;

[0117] S433: Input the matched data set in S42 into the trained random forest model for prediction. The random forest model outputs the optimal control parameters by integrating the prediction results of all decision trees. The specific formula is: ,in, represents the predicted optimal control parameter vector (including the predicted suspension height, hydraulic pressure and operating speed), N is the total number of decision trees, Represents the prediction result of the jth decision tree;

[0118] S434: The random forest model generates the control parameters of the suspension device, including suspension height, hydraulic pressure and operating speed, based on the input data set. The model improves the accuracy and robustness of the prediction by integrating the prediction results of multiple decision trees. Through the above steps, the machine learning algorithm based on random forest can conduct in-depth analysis of the matching data set and accurately predict the suspension control parameters that best match the current operating conditions. The random forest algorithm improves the accuracy of the prediction and the robustness of the system by integrating the prediction results of multiple decision trees, ensuring the effectiveness and applicability of the control strategy. This process not only optimizes the operating efficiency and quality, but also reduces energy consumption and mechanical wear, and enhances the overall performance and reliability of the tractor.

[0119] S5 specifically includes:

[0120] S51: The collaborative operation module receives the preliminary control strategy generated by the intelligent decision module, which includes the suspension height , hydraulic pressure and operating speed Specific parameters of

[0121] S52: transmitting the received preliminary control strategy to each operation unit, including the tillage unit and the sowing unit, through the internal bus. The bus transmission specifically adopts the control area network (CAN) protocol to ensure the real-time and reliability of data transmission;

[0122] S53: Each operation unit receives the preliminary control strategy and adjusts its own operation state according to the strategy parameters, wherein the tillage unit adjusts its depth control and force control to match the suspension height and hydraulic pressure; the sowing unit adjusts its seed release speed and position control to match the operation speed;

[0123] S54: The collaborative operation module monitors the operation status and feedback data of each operation unit in real time through sensors, including actual tillage depth and actual sowing speed;

[0124] S55: The collaborative operation module aggregates the feedback data of each operation unit and compares it with the preliminary control strategy, and adjusts the preliminary control strategy through the built-in collaborative optimization algorithm to ensure the coordination and consistency between the operation units; the collaborative optimization algorithm is implemented through the following steps:

[0125] Step 1: Define the objective function F, which contains the performance indicators of each operation unit, and the expression is:

[0126] ,in, represents the performance function of the i-th job unit, is the corresponding weight coefficient, m is the total number of operating units, and x is the control parameter vector, including , and ;

[0127] Step 2: Use the gradient descent method to optimize the objective function F and calculate the gradient , the formula is: ;

[0128] Step 3: Update the control parameter x according to the gradient, the formula is: ,in, is the learning rate, is the updated control parameter vector.

[0129] S56: According to the adjusted control strategy, dynamically adjust the operating parameters of each work unit to ensure the coordination and consistency of each unit in the collaborative operation; through the above steps, the collaborative operation module can effectively receive the preliminary control strategy generated by the intelligent decision-making module, and interact with other work units for data exchange and collaborative control. Through the collaborative optimization algorithm, dynamically adjust the control strategy according to the real-time feedback data of each work unit to ensure the coordination and consistency between the work units and improve the overall operation efficiency and quality.

[0130] S6 specifically includes:

[0131] S61: The feedback control module receives the real-time collected operation status data from the data acquisition module, including the suspension height , hydraulic system pressure Information and operation speed information;

[0132] S62: The feedback control module performs preliminary processing on the received data to ensure the accuracy and stability of the data. The preliminary processing includes data filtering, denoising and data calibration to eliminate the influence of sensor errors and environmental interference on the data.

[0133] S63: The feedback control module compares the actual operation state data after preliminary processing with the expected parameters in the preliminary control strategy generated by the intelligent decision module to identify the deviation between the operation state and the expected control parameters;

[0134] Calculate the deviation of the suspension height , the formula is: ,in, The expected suspension height generated by the intelligent decision module, is the actual suspension height;

[0135] Calculate the deviation of hydraulic pressure , the formula is: ,in, The expected hydraulic pressure generated by the intelligent decision module, is the actual hydraulic pressure;

[0136] Calculate the deviation of the working speed , the formula is: ,in, The expected operation speed generated for the intelligent decision-making module, is the actual operating speed;

[0137] Through the above steps, the feedback control module can accurately identify the deviation between the actual operating state of the suspension device and the expected control parameters by receiving and processing the operating state data in real time. By calculating the deviation value of each parameter, the system can accurately locate the control parameters that need to be adjusted, thereby ensuring that the operating state of the suspension device is always consistent with the optimal control strategy.

[0138] S7 specifically includes:

[0139] S71: The intelligent decision module receives the deviation data calculated by the feedback control module, including the suspension height deviation , hydraulic pressure deviation and operating speed deviation ;

[0140] S72: The intelligent decision-making module analyzes the deviation data to determine the trend and potential causes of the deviation. The specific analysis includes:

[0141] Conduct trend analysis on deviation data to determine whether the deviation is temporary or persistent;

[0142] Analyze the deviations caused by potential environmental changes and changes in operating conditions based on historical data and current environmental data in the job task library;

[0143] S73: Based on the analysis results, the intelligent decision module uses an adaptive adjustment algorithm to dynamically adjust the preliminary control strategy. Specifically, when the deviation is a persistent deviation, the intelligent decision module will adjust the corresponding control parameters to reduce the deviation; when the deviation is a temporary deviation, the intelligent decision module will monitor and record the deviation data and will not adjust the parameters immediately to prevent over-correction; the dynamic adjustment specifically uses the proportional integral differential (PID) control algorithm to correct the parameters:

[0144] The formula for updating the suspension height control parameters is: ;

[0145] The formula for updating the hydraulic pressure control parameters is: ;

[0146] The formula for updating the operation speed control parameter is: , in, , and are the proportional, integral and differential coefficients of the PID controller respectively;

[0147] S74: transmitting the adjusted control parameters to the collaborative operation module and related operation units to ensure that all operation units operate according to the new control strategy;

[0148] Through the above steps, the intelligent decision-making module can analyze the trend and cause of the deviation based on the deviation data identified by the feedback control module, and use the adaptive adjustment algorithm to dynamically adjust the preliminary control strategy. This dynamic adjustment mechanism can not only respond in a timely manner to environmental changes and operating conditions that may occur during actual operations, ensuring that the operating state of the suspension device is always consistent with the optimal control strategy, but also improve the system's response speed and operating accuracy.

[0149] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0150] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An intelligent control strategy for cooperative operation of an electro-hydraulic suspension device of a high-horsepower tractor, characterized in that: The following steps are involved: S1: Establish an operation task library and define corresponding suspension device control strategy parameters for different types of agricultural operation tasks; S2: Through the data acquisition module, the position information of the suspension device, the pressure information of the hydraulic system, the operation speed, the terrain data and the environmental information during the operation of the tractor are collected in real time and stored in the data storage unit; S3: The data processing module pre-processes the real-time data collected in the data storage unit, including removing noise and abnormal data, and performing feature extraction and data fusion to generate a unified data set; S4: The intelligent decision-making module based on the machine learning algorithm analyzes the data set and generates a preliminary suspension control strategy based on the parameters in the task library, including specific control parameters of suspension height, hydraulic pressure and operation speed; The S4 specifically includes: S41: The intelligent decision module receives the preprocessed and fused data set; S42: Matching the received data set with the preset parameters in the job task library to generate the best matching job task; S43: Analyze the matched data set based on the random forest machine learning algorithm to determine the suspension control parameters that best match the current operating conditions; S44: Based on the analysis results of the random forest algorithm, a preliminary suspension control strategy is generated, which specifies specific parameters of suspension height, hydraulic pressure and operating speed to best adapt to the current operating environment; The S42 specifically includes: S421: Preliminarily matching the preprocessed and fused data set with the preset parameters in the job task library; S422: Calculate the difference between each parameter in the current data set and the preset parameters in the job task library, the formula is: ,in, Represents the difference between the i-th parameter in the current data set and the j-th preset parameter in the job task library. Represents the i-th parameter value in the current data set, Indicates the jth preset parameter value in the job task library; S423: Calculate the total difference of each task to determine the most matching task. The following formula is: ,in, represents the total difference of the jth job task in the job task library, and n represents the total number of parameters; S424: Select the job task with the smallest total difference as the most matching job task at present, the formula is: ,in, Indicates the most matching job task index, that is, the total difference Minimal homework tasks, is the minimum function; S425: Using the preset parameters of the job task that best matches the data set as preliminary control parameters; The S43 specifically includes: S431: constructing a training data set, including historical operation data and corresponding control parameters, the data set including suspension height, hydraulic pressure, operation speed, corresponding environmental conditions and operation results; S432: Use the training data set to train a random forest model. The random forest consists of multiple decision trees, and each decision tree is constructed using the following formula: ,in, represents the decision tree model, X represents the input feature vector, represents the i-th decision tree, represents the corresponding weight, and n is the number of decision trees; S433: Input the matched data set in S42 into the trained random forest model for prediction. The random forest model outputs the optimal control parameters by integrating the prediction results of all decision trees. The specific formula is: ,in, represents the predicted optimal control parameter vector, N is the total number of decision trees, Represents the prediction result of the jth decision tree; S434: The random forest model generates control parameters of the suspension device according to the input data set, including suspension height, hydraulic pressure and operating speed; S5: The collaborative operation module receives the preliminary control strategy generated by the intelligent decision-making module, and performs data exchange and collaborative control with other operation units to ensure coordination and consistency among the operation units; S6: The feedback control module receives the operating state data collected in real time from the data acquisition module, processes and analyzes the data, and identifies the deviation between the operating state and the expected control parameters; S7: The intelligent decision-making module dynamically adjusts the preliminary control strategy based on the deviations identified by the feedback control module to cope with environmental changes and operating condition changes that occur during the actual operation process.

2. The intelligent control strategy for cooperative operation of the electro-hydraulic suspension device of a high-horsepower tractor according to claim 1 is characterized in that: The S1 specifically includes: S11: Classify different types of agricultural tasks, including tillage, sowing, and harvesting tasks; S12: For each operation task, collect relevant operation parameters and historical data, including information on terrain characteristics, soil type, operation depth, operation speed, and operation tools, and organize and standardize the data to form a database; S13: Based on the collected historical data and relevant operating parameters, the optimal suspension control strategy parameters for various operating tasks are determined using statistical analysis and machine learning algorithm analysis, including specific parameters of suspension height, hydraulic pressure and operating speed; S14: The control strategy parameters obtained through analysis are stored in the operation task library, and a corresponding relationship between the parameters and specific operation tasks is established for use in the actual operation process.

3. The intelligent control strategy for cooperative operation of the electro-hydraulic suspension device of a high-horsepower tractor according to claim 1 is characterized in that: The S2 specifically includes: S21: The data acquisition module is equipped with a variety of sensors, including a position sensor, a pressure sensor, a speed sensor, a terrain scanner and an environmental monitor; wherein the position sensor is used to monitor the position of the suspension device in real time; the pressure sensor is used to monitor the working pressure of the hydraulic system; the speed sensor is used to record the operating speed of the tractor; the terrain scanner is used to obtain the terrain data of the current operating area; and the environmental monitor is used to collect environmental information such as temperature and humidity; S22: The data collected by each sensor is transmitted to the central processing unit in real time through the data collection unit, and the central processing unit is used to perform preliminary time stamp marking and classification processing on the received data; S23: During the data collection process, implement data filtering algorithms to remove noise or abnormal data, the filtering algorithms include signal smoothing, outlier detection and data normalization; S24: storing the preliminarily processed and verified data in a data storage unit; S25: Establish data index. Each data item is marked with a unique index and a label of the relevant job task, so that the data can be quickly associated with the scheduled job task.

4. The intelligent control strategy for cooperative operation of the electro-hydraulic suspension device of a high-horsepower tractor according to claim 1 is characterized in that: The S3 specifically includes: S31: extracting real-time data from the data storage unit, including position information of the suspension device, pressure information of the hydraulic system, operating speed, terrain data and environmental information; S32: applying a data cleaning algorithm to remove noise and abnormal data from the extracted data, specifically using a median filter algorithm and a mean filter algorithm to smooth the data and eliminate instantaneous noise; and using a statistical analysis method to detect and remove abnormal values; S33: Extract features by analyzing the cleaned data to extract features, including the displacement change rate of the suspension device; the fluctuation range of the hydraulic system pressure; the instantaneous speed and acceleration of the tractor; the terrain slope and surface roughness; and the change trend of the ambient temperature and humidity; S34: performing data fusion, specifically using a Kalman filter algorithm to fuse multi-sensor data, and performing dimensionality reduction processing on multi-dimensional feature data to reduce redundant data, so as to form a comprehensive feature data set; S35: Generate a unified data set, store the fused comprehensive feature data set in a data storage unit, and add a timestamp and a task label to each data set.

5. The intelligent control strategy for cooperative operation of the electro-hydraulic suspension device of a high-horsepower tractor according to claim 1 is characterized in that: The S5 specifically includes: S51: The collaborative operation module receives the preliminary control strategy generated by the intelligent decision module, which includes the suspension height , hydraulic pressure and operating speed Specific parameters of S52: transmitting the received preliminary control strategy to each operation unit, including the tillage unit and the sowing unit, through the internal bus; S53: Each operating unit receives the preliminary control strategy and adjusts its own operating state according to the strategy parameters; S54: The collaborative operation module monitors the operation status and feedback data of each operation unit in real time through sensors, including actual tillage depth and actual sowing speed; S55: The collaborative operation module aggregates the feedback data of each operation unit and compares it with the preliminary control strategy, and adjusts the preliminary control strategy through the built-in collaborative optimization algorithm to ensure coordination and consistency among the operation units; S56: According to the adjusted control strategy, the operating parameters of each operating unit are dynamically adjusted to ensure the coordination and consistency of each unit in the collaborative operation.

6. The intelligent control strategy for cooperative operation of the electro-hydraulic suspension device of a high-horsepower tractor according to claim 1 is characterized in that: The S6 specifically includes: S61: The feedback control module receives the real-time collected operation status data from the data acquisition module, including the suspension height , hydraulic system pressure Information and operation speed information; S62: The feedback control module performs preliminary processing on the received data, and the preliminary processing includes data filtering, denoising and data calibration; S63: The feedback control module compares the actual operation state data after preliminary processing with the expected parameters in the preliminary control strategy generated by the intelligent decision module to identify the deviation between the operation state and the expected control parameters; Calculate the deviation of the suspension height , the formula is: ,in, The expected suspension height generated by the intelligent decision module, is the actual suspension height; Calculate the deviation of hydraulic pressure , the formula is: ,in, The expected hydraulic pressure generated by the intelligent decision module, is the actual hydraulic pressure; Calculate the deviation of the working speed , the formula is: ,in, The expected operation speed generated for the intelligent decision-making module, The actual operating speed.

7. The intelligent control strategy for cooperative operation of the electro-hydraulic suspension device of a high-horsepower tractor according to claim 1 is characterized in that: The S7 specifically includes: S71: The intelligent decision module receives the deviation data calculated by the feedback control module, including the suspension height deviation, the hydraulic pressure deviation and the operating speed deviation; S72: The intelligent decision-making module analyzes the deviation data to determine the trend and potential causes of the deviation; S73: Based on the analysis results, the intelligent decision module uses an adaptive adjustment algorithm to dynamically adjust the preliminary control strategy. Specifically, when the deviation is a persistent deviation, the intelligent decision module will adjust the corresponding control parameters to reduce the deviation; when the deviation is a temporary deviation, the intelligent decision module will monitor and record the deviation data and will not adjust the parameters immediately; the dynamic adjustment specifically uses the proportional integral differential control algorithm to correct the parameters: The formula for updating the suspension height control parameters is: ; The formula for updating the hydraulic pressure control parameters is: ; The formula for updating the operation speed control parameter is: ,in, , and are the proportional, integral and differential coefficients of the PID controller respectively; S74: The adjusted control parameters are transmitted to the collaborative operation module and related operation units to ensure that all operation units operate according to the new control strategy.

Citation Information

Patent Citations

  • Intelligent control method and system for intelligent flow film composting process

    CN112062617A

  • Methods and systems for data collection, learning, and streaming of machine signals for analytics and maintenance using the industrial internet of things

    CN112703457A