Intelligent control model and method for cleaning of rice and wheat combine harvester

By collecting and analyzing the operation and environmental data of the rice and wheat combined harvester, using fuzzy reasoning and intelligent control strategies, dynamically adjusting the cleaning parameters, solving the problem that traditional methods are difficult to cope with complex environments, improving the cleaning efficiency and stability, and realizing the Higher operating speed and productivity.

CN119024749APending Publication Date: 2024-11-26SUZHOU ZHISHOU AGRICULTURAL TECHNOLOGY CO LTD
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411140731.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-20
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

The cleaning control method of traditional rice and wheat combined harvesters is difficult to cope with complex nonlinear systems and variable operating environments, resulting in unstable cleaning and high grain loss rate.

Method used

The intelligent control method for cleaning and selection of rice and wheat combined harvester is adopted. By collecting operation data and environmental data, linear regression analysis and fuzzy reasoning are performed, control strategies are generated, and clear selection parameters are dynamically adjusted. Combined with PID control or neural network control, it is adapted to complex environments.

Benefits of technology

It improves the stability and accuracy of the cleaning process, reduces grain loss and impurity residues, and automatically adjusts reduces human intervention, improves operating efficiency and productivity, and adapts to different crop varieties and environmental conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119024749A_ABST
    Figure CN119024749A_ABST
Patent Text Reader

Abstract

The invention discloses a rice and wheat combine harvester cleaning intelligent regulation and control model and method, and particularly relates to the technical field of agricultural machinery intelligent regulation and control, and the method comprises the following steps: firstly, collecting operation data and environment data, carrying out linear regression analysis to obtain decision coefficients of cleaning variables, and summarizing and calculating the decision coefficients as a comprehensive measurement linear index; after the environmental data is subjected to standardization processing, the volatility, randomness and fuzziness of environmental variables are calculated and summarized into an environmental comprehensive volatility index; the two indexes serve as input through fuzzy reasoning, a control strategy is generated, parameter regulation and control are conducted on cleaning variables according to the strategy, and the intelligent and efficient cleaning process is achieved; according to the method, fuzzy logic and an intelligent control algorithm are combined, a control strategy can be automatically adjusted according to real-time operation conditions, and when an operation environment changes or system characteristics are nonlinear, flexible response can be achieved through a neural network control strategy, and it is ensured that the agricultural machinery is always in the optimal working state.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of agricultural machinery, and more specifically, to an intelligent control model and method for cleaning a rice and wheat combine harvester. Background Art

[0002] In agricultural production, rice-wheat combine harvesters are important equipment in modern agriculture, and their efficiency and operation quality directly affect the harvesting effect and economic benefits of crops. The cleaning process is one of the key links in the work of the combine harvester, involving operations such as separation and screening of grains and impurities. However, due to the characteristics of rice and wheat crops and the complex and changeable working environment, traditional cleaning control methods are often difficult to cope with the changing working conditions and environment, resulting in unstable cleaning effects and high grain loss rates. Traditional control strategies mainly rely on fixed-parameter PID control or manual adjustment based on experience, but these methods are usually difficult to adapt to complex nonlinear systems and changeable working environments. For example, differences in rice and wheat varieties, changes in humidity, fluctuations in the external environment, and other factors will affect the cleaning effect. If the cleaning parameters cannot be adjusted effectively in real time, it may lead to incomplete cleaning, increased grain loss, or reduced equipment efficiency. Therefore, a rice-wheat combine harvester cleaning intelligent control model and method are proposed to solve the above problems. Summary of the invention

[0003] To achieve the above object, the present invention provides the following technical solutions:

[0004] A method for intelligently controlling the cleaning of a rice and wheat combine harvester comprises the following steps:

[0005] Collect the operation data of the rice and wheat cleaning by the combine harvester, as well as the environmental data during the rice and wheat cleaning by the combine harvester;

[0006] Perform linear regression analysis between each cleaning variable in the operation data and the cleaning effect, obtain the determination coefficient of each cleaning variable, and then summarize and calculate the determination coefficients of all cleaning variables;

[0007] Standardize the environmental variables in the environmental data and calculate the volatility, randomness and fuzziness of each environmental variable respectively, and then calculate the volatility, randomness and fuzziness of all environmental variables in aggregate;

[0008] Fuzzy reasoning is used to reason the results of the determination coefficient summary calculation of all the selected variables, the volatility, randomness and fuzziness summary calculation results of all environmental variables, and generate the output of the control strategy;

[0009] The control strategy output by fuzzy reasoning is selected for execution, and parameters of the cleaning variables in the rice and wheat cleaning operation data are adjusted.

[0010] In a preferred embodiment, the control strategy refers to PID control or neural network control, and the neural network control is pre-trained.

[0011] In a preferred embodiment, the operation data includes a plurality of preset cleaning variable data, such as screen vibration frequency, fan speed, grain flow rate, initial impurity content of grain and other related data.

[0012] In a preferred embodiment, the environmental data includes a plurality of preset environmental variable data, such as crop humidity, crop variety, crop harvesting environment temperature and other related data.

[0013] In a preferred embodiment, a linear regression analysis is performed between each cleaning variable in the operation data and the cleaning effect, the determination coefficient of each cleaning variable is obtained, and then the determination coefficients of all cleaning variables are aggregated and calculated, which means:

[0014] Each cleaning variable is set as the independent variable Xi, and the dependent variable is set as Y, and the following linear regression model is established: Y = ai*Xi+bi+∈i; Y represents the cleaning effect, ai and bi represent the regression coefficients corresponding to the cleaning variable number i, and ∈i is the error term corresponding to the cleaning variable number i;

[0015] For the target clearing variable, the total sum of squares of the total variation of the dependent variable Y relative to its mean is calculated and expressed as SST, the sum of squares of the residuals that cannot be explained in the linear regression model of the target clearing variable is calculated and expressed as SSE, and then the proportion of the variation explained by the model to the total variation Ri is calculated 2 , the calculation formula is:

[0016] Get the determination coefficient Ri corresponding to all the selected variables 2 , and then substitute it into the following calculation formula:

[0017] ri is the influence coefficient corresponding to the cleaning variable number i, LL represents the comprehensive linear index of all cleaning variables, and n represents the total number of cleaning variables.

[0018] In a preferred embodiment, the environmental variables in the environmental data are standardized and the volatility, randomness and fuzziness of each environmental variable are calculated respectively, and then the volatility, randomness and fuzziness of all environmental variables are calculated in aggregate, which means:

[0019] First, the environmental variables in the environmental data are standardized and marked as Ei, where i represents the number of the environmental variable. For each standardized environmental variable Ei, its volatility, i.e., standard deviation σ(Ei), is calculated, and then its randomness, i.e., variance Var(Ei), is calculated. Then, the triangle membership function is used to calculate the fuzziness. The calculation formula is: a, b, c are all parameters of the membership function, μ(Ei) represents the fuzziness corresponding to the environmental variable number i;

[0020] The weighted geometric mean method is used to summarize the volatility, randomness and fuzziness of all environmental variables. The calculation formula is:

[0021] k represents the total number of environmental variables, EFI represents the comprehensive environmental fluctuation index, and wi represents the proportional coefficient corresponding to the environmental variable number i.

[0022] In a preferred embodiment, fuzzy reasoning is used to reason the results of the determination coefficient summary calculation of all the selected variables, the volatility, randomness and fuzziness summary calculation results of all the environmental variables, and the output of the control strategy is generated:

[0023] The comprehensive linear index calculated by summarizing the coefficient of determination of all selected variables and the comprehensive environmental volatility index calculated by summarizing the volatility, randomness and fuzziness of all environmental variables are obtained and used as input variables together. The type of control strategy is used as the output variable. The input variables are fuzzified and the values ​​of the input variables are converted into fuzzy sets. The output variables are fuzzified and the output variables are converted into fuzzy sets. Fuzzy rules are formulated to describe the control requirements under different data type combinations. The fuzzified input variables are inferred through fuzzy rules to determine the type of control strategy to be used.

[0024] In a preferred embodiment, a rice and wheat combine harvester cleaning intelligent control model includes:

[0025] A data collection module is used to collect the operation data of the rice and wheat cleaning of the combine harvester, as well as the environmental data of the rice and wheat cleaning of the combine harvester;

[0026] The cleaning variable linear analysis module is used to perform linear regression analysis between each cleaning variable in the operation data and the cleaning effect, obtain the determination coefficient of each cleaning variable, and then summarize and calculate the determination coefficients of all cleaning variables;

[0027] Environmental variable evaluation module, used to standardize environmental variables in environmental data and calculate the volatility, randomness and fuzziness of each environmental variable respectively, and then aggregate the volatility, randomness and fuzziness of all environmental variables;

[0028] The fuzzy reasoning module is used to use fuzzy reasoning to reason the results of the determination coefficient summary calculation of all the selected variables and the results of the volatility, randomness and fuzziness summary calculation of all the environmental variables to generate the output of the control strategy;

[0029] The control strategy execution module is used to select the control strategy output by fuzzy reasoning for execution and to perform parameter control on the cleaning variables in the rice and wheat cleaning operation data.

[0030] Technical effects and advantages of the present invention:

[0031] The present invention can adapt to complex and changeable working environments by comprehensively weighing the calculation of linear index and comprehensive environmental fluctuation index, and dynamically selecting control strategies using fuzzy reasoning, thereby effectively improving the stability and accuracy of the rice and wheat combine harvester cleaning process, and reducing grain loss and impurity residues. The present invention combines fuzzy logic and intelligent control algorithms, and can automatically adjust the control strategy according to real-time working conditions. When the working environment changes or the system characteristics are nonlinear, the present invention can flexibly respond through a neural network control strategy to ensure that the agricultural machinery is always in the best working state.

[0032] Traditional cleaning control methods often require operators to perform manual adjustments and monitoring, while the intelligent control method of the present invention can automatically perform real-time adjustments based on operating data and environmental data, greatly reducing the need for human intervention and improving the convenience and efficiency of operation. The present invention rapidly generates and executes the optimal control strategy through real-time data acquisition and analysis, significantly improving the operating efficiency of the rice and wheat combine harvester. By reducing unnecessary adjustment time and avoiding inefficient operations, the present invention can achieve higher operating speeds and productivity while ensuring the quality of cleaning. Whether in different crop varieties, different humidity conditions, or in various complex external environments, the present invention can adapt to these changes and optimize the cleaning process through an intelligent control mechanism. This makes the present invention widely applicable and able to meet the diverse needs of agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0034] Figure 1 The present invention is a schematic diagram of an intelligent control method for cleaning a rice and wheat combine harvester.

[0035] Figure 2 The present invention is a schematic diagram of an intelligent control model for cleaning a rice and wheat combine harvester. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] Reference Figure 1 - Figure 2 The following examples are obtained:

[0038] Embodiment 1: In agricultural production, rice and wheat combine harvesters are important equipment for modern agriculture, and their efficiency and operation quality directly affect the harvesting effect and economic benefits of crops. The cleaning process is one of the key links in the work of the combine harvester, involving operations such as separation and screening of grains and impurities. However, due to the characteristics of rice and wheat crops and the complexity and changeability of the operating environment, traditional cleaning control methods are often difficult to cope with the changeable working conditions and environment, resulting in unstable cleaning effects and high grain loss rates. Traditional control strategies mainly rely on fixed parameter PID control or manual adjustment based on experience, but these methods are usually difficult to adapt to complex nonlinear systems and changeable operating environments. For example, factors such as differences in rice and wheat varieties, changes in humidity, and fluctuations in the external environment will affect the cleaning effect. If the cleaning parameters cannot be adjusted effectively in real time, incomplete cleaning, increased grain loss or reduced equipment working efficiency may result. Therefore, the present invention proposes an intelligent control method for cleaning a rice and wheat combine harvester, comprising the following steps: collecting operation data of rice and wheat cleaning by the combine harvester, and environmental data of rice and wheat cleaning by the combine harvester; performing linear regression analysis between each cleaning variable and the cleaning effect in the operation data, obtaining the determination coefficient of each cleaning variable, and then summarizing and calculating the determination coefficients of all cleaning variables; standardizing the environmental variables in the environmental data and calculating the volatility, randomness and fuzziness of each environmental variable respectively, and then summarizing and calculating the volatility, randomness and fuzziness of all environmental variables; using fuzzy reasoning to infer the results of the summary calculation of the determination coefficients of all cleaning variables and the results of the summary calculation of the volatility, randomness and fuzziness of all environmental variables, and generating the output of the control strategy; selecting the control strategy output by the fuzzy reasoning for execution, and performing parameter control on the cleaning variables in the operation data of rice and wheat cleaning.

[0039] The method is proposed based on the following considerations: Challenges of nonlinear systems: There are complex nonlinear relationships between multiple variables (such as screen vibration frequency, fan speed, etc.) in the rice and wheat cleaning process and the cleaning effect. Traditional linear control methods (such as PID control) are often unable to cope with these nonlinear problems, so it is necessary to introduce more flexible control strategies, such as the pre-trained neural network control mentioned in the present invention, to adapt to these complex system characteristics.

[0040] Impact of environmental uncertainty: Multiple factors in the operating environment (such as crop humidity, temperature, wind speed, etc.) have large fluctuations and uncertainties. These environmental factors have a significant impact on the cleaning effect, and a method is needed to evaluate environmental changes in real time and dynamically adjust the control strategy to cope with random fluctuations in the environment.

[0041] Demand for intelligent control: Modern agricultural machinery is developing towards intelligence. Intelligent control can not only improve work efficiency, but also reduce manual intervention and reduce the difficulty of operation. By collecting data in real time, analyzing data and automatically generating control strategies, the cleaning process can be intelligentized, thereby improving the cleaning quality and equipment utilization.

[0042] The control strategy refers to PID control or neural network control, and the neural network control is pre-trained.

[0043] Implementation steps of PID control: First, clarify the target values ​​of the cleaning variables (such as screen vibration frequency, fan speed, etc.) that need to be controlled. These target values ​​are usually determined based on the best cleaning effect (such as the lowest impurity content or the highest grain integrity rate). The cleaning effect in the present invention can be the impurity content; during the cleaning operation, data related to the cleaning variables are collected in real time. These data include the current cleaning variable value (actual value) and the cleaning effect. The difference between the actual value and the target value of the cleaning variable is calculated, that is, the error. The error value is an important basis for the adjustment of the PID controller. According to the calculated error, the PID controller is adjusted through three parts: proportional control (P): directly adjust the output of the control variable according to the size of the error. The larger the error, the larger the adjustment range. Integral control (I): Considering the cumulative effect of the error over time, further adjust the control output to eliminate long-term small errors. Differential control (D): Considering the rate of change of the error, predicting the future error trend, and pre-adjusting to avoid drastic fluctuations in the error. Through the combined effect of these three parts, the PID controller adjusts the parameters of the cleaning variable so that it gradually approaches the target value, and finally achieves a stable cleaning effect. The cleaning effect is continuously monitored and fed back to the PID controller in real time. If new errors occur, the PID controller will recalculate and adjust according to the new errors to ensure that the cleaning effect remains at the best state.

[0044] Training steps for neural network control: Collect a large amount of historical data on the cleaning operation of rice and wheat combine harvesters. These data include cleaning variables (such as screen vibration frequency, fan speed, etc.), environmental variables (such as crop humidity, temperature, etc.) and corresponding cleaning effects (such as impurity content, grain integrity, etc.). Standardize or normalize the collected data to eliminate the impact of dimensions, and handle outliers and missing data to ensure data quality and consistency. According to the complexity of the cleaning process, select a suitable neural network architecture. Commonly used architectures include feedforward neural networks (FNN) or convolutional neural networks (CNN). Parameters such as the number of network layers and the number of neurons in each layer are set according to the needs of the task. Divide the preprocessed data into training sets and test sets, usually 80% of the data is used for training and 20% of the data is used for testing. The training set is used to train the neural network, and the test set is used to evaluate the training effect. Use the training set data to train the neural network through the back propagation algorithm. During the training process, the network adjusts the weights and biases so that the output gradually approaches the target value and the error gradually decreases. During the training process, parameters such as learning rate and number of iterations are set to ensure that the neural network can fully learn the data features without overfitting. The trained neural network is verified using the test set data to evaluate its performance on unseen data. Once the performance reaches the preset standard, it can be deployed and the trained neural network model is deployed in the cleaning control system of the rice and wheat combine harvester. The model will automatically generate a control strategy suitable for the current operating conditions based on the real-time collected operating data, adjust the cleaning variables in real time, and optimize the cleaning effect.

[0045] It should be noted that the operation data includes multiple preset cleaning variable data, such as screen vibration frequency, fan speed, grain flow rate, initial impurity content of grains, etc. Environmental data includes multiple preset environmental variable data, such as crop humidity, crop variety, crop harvesting environment temperature, etc.

[0046] Perform linear regression analysis between each cleaning variable and the cleaning effect in the operation data to obtain the determination coefficient of each cleaning variable, and then summarize and calculate the determination coefficients of all cleaning variables: each cleaning variable is set as the independent variable Xi, the dependent variable is set as Y, and the following linear regression model is established: Y = ai*Xi+bi+∈i; Y represents the cleaning effect, ai and bi represent the regression coefficients corresponding to the cleaning variable number i, and ∈i is the error term corresponding to the cleaning variable number i;

[0047] For the target cleanup variable, the total sum of squares of the total variation of the dependent variable Y relative to its mean is calculated and expressed as SST. m represents the total number of data samples of the target selection variable, Y j represents the jth observation value, The data mean of the target cleaning variable is represented by the residual sum of squares that cannot be explained in the linear regression model of the target cleaning variable and expressed as SSE. SSE = Represents the jth predicted value of the target variable, and then calculates the proportion of the variation explained by the model to the total variation Ri 2 , the calculation formula is: Ri 2 It is the determination coefficient between each cleaning variable and the cleaning effect, which is used to measure the explanatory power of this variable on the cleaning effect. Its value ranges from 0 to 1. The closer it is to 1, the stronger the explanatory power of the cleaning variable on the cleaning effect; the closer it is to 0, the weaker the explanatory power.

[0048] Get the determination coefficient Ri corresponding to all the selected variables 2 , and then substitute it into the following calculation formula:

[0049] ri is the influence coefficient corresponding to the cleaning variable number i, which reflects the importance or influence of each cleaning variable in the overall cleaning process. The larger the weight, the more significant the influence of the variable on the cleaning effect. n represents the total number of cleaning variables. LL represents the comprehensive linear index of all cleaning variables. The value reflects the linearity of the entire system by weighted average calculation of the determination coefficient of all cleaning variables and their corresponding importance weights. The larger the LL, the higher the linearity of the overall system. In other words, the relationship between each cleaning variable and the cleaning effect in the system is closer to linearity, and the variable has a stronger ability to explain the effect. This means that in this case, the use of linear control strategies (such as PID control) may be more effective and applicable. If the LL value is very high, it means that most of the cleaning variables in the system can effectively explain the cleaning effect through a linear model, the system has good predictability, and the linear control method can adapt well to changes in the operation process. On the contrary, if the LL value is low, it means that the system has strong nonlinear characteristics, the linear model has weak explanatory power, and more complex control methods (such as neural network control) may be required to handle it.

[0050] Standardizing the environmental variables in the environmental data and calculating the volatility, randomness, and fuzziness of each environmental variable separately, and then aggregating the volatility, randomness, and fuzziness of all environmental variables means:

[0051] First, the environmental variables in the environmental data are standardized and marked as Ei, where i represents the number of the environmental variable. For each standardized environmental variable Ei, its volatility, i.e., standard deviation σ(Ei), is calculated, and then its randomness, i.e., variance Var(Ei), is calculated. The standard deviation and variance formulas are not repeated here. Then, the triangle membership function is used to calculate the fuzziness. The calculation formula is: a, b, c are all parameters of the membership function, μ(Ei) represents the fuzziness corresponding to the environmental variable number i; the standard deviation σ(Ei) is the volatility of the environmental variable Ei, reflecting the discreteness of the environmental variable value. The larger the standard deviation, the greater the volatility of the environmental variable, that is, the greater the value variation of the variable at different time points or locations. The variance Var(Ei) is the randomness of the environmental variable Ei, reflecting the uncertainty or unpredictability of the environmental variable. The larger the variance, the more unstable the value of the environmental variable and the stronger the randomness. The fuzziness reflects the uncertainty of the variable within a specific range. The higher the fuzziness value, the worse the certainty of the variable, that is, the larger the fluctuation range under given conditions.

[0052] The weighted geometric mean method is used to summarize the volatility, randomness and fuzziness of all environmental variables. The calculation formula is:

[0053] k represents the total number of environmental variables, wi represents the proportional coefficient corresponding to the environmental variable number i. The larger the proportional coefficient, the greater the contribution of the variable to the environmental fluctuation index, that is, the variable has a higher importance in the comprehensive evaluation. EFI represents the comprehensive environmental fluctuation index, which is used to measure the volatility and uncertainty of the overall working environment. The larger the EFI, the higher the volatility, randomness and fuzziness of the overall working environment. This means that the environmental conditions are unstable, change dramatically, and have greater uncertainty. High EFI usually indicates that the working environment is complex and difficult to predict, which may have a significant impact on the cleaning process. Therefore, in this case, more complex control strategies (such as neural network control) may be required to adapt to environmental changes and maintain a stable cleaning effect.

[0054] Fuzzy reasoning is used to reason the results of the determination coefficient summary calculation of all selected variables, the volatility, randomness and fuzziness summary calculation results of all environmental variables, and the output of the control strategy is generated:

[0055] The comprehensive linear index calculated by summarizing the coefficient of determination of all selected variables and the comprehensive environmental volatility index calculated by summarizing the volatility, randomness and fuzziness of all environmental variables are obtained and used as input variables together. The type of control strategy is used as the output variable. The input variables are fuzzified and the values ​​of the input variables are converted into fuzzy sets. The output variables are fuzzified and the output variables are converted into fuzzy sets. Fuzzy rules are formulated to describe the control requirements under different data type combinations. The fuzzified input variables are inferred through fuzzy rules to determine the type of control strategy to be used.

[0056] Specifically, in the intelligent control process of the rice-wheat combine harvester, two key input variables need to be obtained first: the comprehensive linear index and the comprehensive environmental fluctuation index. The comprehensive linear index is calculated by summarizing the determination coefficients of all cleaning variables, which reflects the linearity of the entire system. The comprehensive environmental fluctuation index is calculated by summarizing the volatility, randomness and fuzziness of all environmental variables, representing the overall uncertainty and volatility of the operating environment.

[0057] After determining the two input variables, the two variables are used as input and the type of control strategy is used as the output variable. In order to make effective judgments in the fuzzy reasoning process, the two input variables need to be fuzzified. Specifically, the values ​​of the comprehensive linear index and the environmental comprehensive volatility index are converted into fuzzy sets. For example, the comprehensive linear index can be fuzzified into three levels: high, medium, and low, while the environmental comprehensive volatility index can be fuzzified into three levels: stable, medium, and volatile.

[0058] At the same time, the output variables of the control strategy are also fuzzified, that is, the type of control strategy is converted into a fuzzy set. The control strategy type can include different control algorithms, such as traditional proportional integral differential control, neural network control or other intelligent control methods.

[0059] Once both the input and output variables are fuzzified, the next step is to formulate fuzzy rules. These rules define the control strategies that should be adopted under different data combinations. For example, when the comprehensive linear index is high and the environmental comprehensive volatility index is low, the proportional integral derivative control strategy may be more suitable; when the comprehensive linear index is low and the environmental comprehensive volatility index is high, the neural network control strategy may be required.

[0060] Finally, the fuzzified input variables are inferred through the established fuzzy rules. This process is to match the fuzzy values ​​of the two input variables with the fuzzy rules to determine the type of control strategy that best suits the current operating conditions. In this way, the optimal control strategy can be automatically selected and executed based on the changes in real-time cleaning variables and environmental variables to ensure the best cleaning effect.

[0061] Embodiment 2: A rice and wheat combine harvester cleaning intelligent control model, comprising:

[0062] A data collection module is used to collect the operation data of the rice and wheat cleaning of the combine harvester, as well as the environmental data of the rice and wheat cleaning of the combine harvester;

[0063] The cleaning variable linear analysis module is used to perform linear regression analysis between each cleaning variable in the operation data and the cleaning effect, obtain the determination coefficient of each cleaning variable, and then summarize and calculate the determination coefficients of all cleaning variables;

[0064] Environmental variable evaluation module, used to standardize environmental variables in environmental data and calculate the volatility, randomness and fuzziness of each environmental variable respectively, and then aggregate the volatility, randomness and fuzziness of all environmental variables;

[0065] The fuzzy reasoning module is used to use fuzzy reasoning to reason the results of the determination coefficient summary calculation of all the selected variables and the results of the volatility, randomness and fuzziness summary calculation of all the environmental variables to generate the output of the control strategy;

[0066] The control strategy execution module is used to select the control strategy output by fuzzy reasoning for execution and to perform parameter control on the cleaning variables in the rice and wheat cleaning operation data.

[0067] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by software simulation of a large amount of data. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0068] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0069] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0070] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0071] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for intelligently controlling the cleaning of a rice and wheat combine harvester, characterized in that: The following steps are involved: Collect the operation data of the rice and wheat cleaning by the combine harvester, as well as the environmental data during the rice and wheat cleaning by the combine harvester; Perform linear regression analysis between each cleaning variable in the operation data and the cleaning effect, obtain the determination coefficient of each cleaning variable, and then summarize and calculate the determination coefficients of all cleaning variables; Standardize the environmental variables in the environmental data and calculate the volatility, randomness and fuzziness of each environmental variable respectively, and then calculate the volatility, randomness and fuzziness of all environmental variables in aggregate; Fuzzy reasoning is used to reason the results of the determination coefficient summary calculation of all the selected variables, the volatility, randomness and fuzziness summary calculation results of all environmental variables, and generate the output of the control strategy; The control strategy output by fuzzy reasoning is selected for execution, and parameters of the cleaning variables in the rice and wheat cleaning operation data are adjusted.

2. The intelligent control method for cleaning a rice and wheat combine harvester according to claim 1, characterized in that: The control strategy refers to PID control or neural network control, and the neural network control is pre-trained.

3. The intelligent control method for cleaning a rice and wheat combine harvester according to claim 2, characterized in that: The job data includes preset multiple cleaning variable data.

4. The intelligent control method for cleaning a rice and wheat combine harvester according to claim 3, characterized in that: The environmental data includes multiple preset environmental variable data.

5. The intelligent control method for cleaning a rice and wheat combine harvester according to claim 3, characterized in that: Perform linear regression analysis on each cleaning variable in the operation data and the cleaning effect, obtain the determination coefficient of each cleaning variable, and then summarize and calculate the determination coefficients of all cleaning variables: Each cleaning variable is set as the independent variable Xi, and the dependent variable is set as Y, and the following linear regression model is established: Y = ai*Xi+bi+∈i; Y represents the cleaning effect, ai and bi represent the regression coefficients corresponding to the cleaning variable number i, and ∈i is the error term corresponding to the cleaning variable number i; For the target clearing variable, the total sum of squares of the total variation of the dependent variable Y relative to its mean is calculated and expressed as SST, the sum of squares of the residuals that cannot be explained in the linear regression model of the target clearing variable is calculated and expressed as SSE, and then the proportion of the variation explained by the model to the total variation Ri is calculated 2 , the calculation formula is: Get the determination coefficient Ri corresponding to all the selected variables 2 , and then substitute it into the following calculation formula: ri is the influence coefficient corresponding to the cleaning variable number i, LL represents the comprehensive linear index of all cleaning variables, and n represents the total number of cleaning variables.

6. The intelligent control method for cleaning a rice and wheat combine harvester according to claim 5, characterized in that: Standardizing the environmental variables in the environmental data and calculating the volatility, randomness, and fuzziness of each environmental variable separately, and then aggregating the volatility, randomness, and fuzziness of all environmental variables means: First, the environmental variables in the environmental data are standardized and marked as Ei, where i represents the number of the environmental variable. For each standardized environmental variable Ei, its volatility, i.e., standard deviation σ(Ei), is calculated, and then its randomness, i.e., variance Var(Ei), is calculated. Then, the triangle membership function is used to calculate the fuzziness. The calculation formula is: a, b, c are all parameters of the membership function, μ(Ei) represents the fuzziness corresponding to the environmental variable number i; The weighted geometric mean method is used to summarize the volatility, randomness and fuzziness of all environmental variables. The calculation formula is: k represents the total number of environmental variables, EFI represents the comprehensive environmental fluctuation index, and wi represents the proportional coefficient corresponding to the environmental variable number i.

7. The intelligent control method for cleaning a rice and wheat combine harvester according to claim 6, characterized in that: Fuzzy reasoning is used to reason the results of the determination coefficient summary calculation of all selected variables, the volatility, randomness and fuzziness summary calculation results of all environmental variables, and the output of the control strategy is generated: The comprehensive linear index calculated by summarizing the coefficient of determination of all selected variables and the comprehensive environmental volatility index calculated by summarizing the volatility, randomness and fuzziness of all environmental variables are obtained and used as input variables together. The type of control strategy is used as the output variable. The input variables are fuzzified and the values ​​of the input variables are converted into fuzzy sets. The output variables are fuzzified and the output variables are converted into fuzzy sets. Fuzzy rules are formulated to describe the control requirements under different data type combinations. The fuzzified input variables are inferred through fuzzy rules to determine the type of control strategy to be used.

8. An intelligent control model for cleaning a rice-wheat combine harvester, based on an intelligent control method for cleaning a rice-wheat combine harvester according to any one of claims 1 to 7, characterized in that: include: A data collection module is used to collect the operation data of the rice and wheat cleaning of the combine harvester, as well as the environmental data of the rice and wheat cleaning of the combine harvester; The cleaning variable linear analysis module is used to perform linear regression analysis between each cleaning variable in the operation data and the cleaning effect, obtain the determination coefficient of each cleaning variable, and then summarize and calculate the determination coefficients of all cleaning variables; Environmental variable evaluation module, used to standardize environmental variables in environmental data and calculate the volatility, randomness and fuzziness of each environmental variable respectively, and then aggregate the volatility, randomness and fuzziness of all environmental variables; The fuzzy reasoning module is used to use fuzzy reasoning to reason the results of the determination coefficient summary calculation of all the selected variables and the results of the volatility, randomness and fuzziness summary calculation of all the environmental variables to generate the output of the control strategy; The control strategy execution module is used to select the control strategy output by fuzzy reasoning for execution and to perform parameter control on the cleaning variables in the rice and wheat cleaning operation data.

Citation Information

Cited By

  • Self-adaptive control method and system for underwater operation power of dredging robot

    CN120010231A

  • Rice and wheat combine harvester cleaning device intelligent regulation and control method based on rice and wheat attributes

    CN120167218A

  • Intelligent dynamic regulation and control method for cleaning operation of rice and wheat combine harvester

    CN120787623A