A self-adaptive control method for a harvester header

Abstract

The application provides a harvester header self-adaptive regulation method, relates to the intelligent technology field of agricultural machinery, and aims at the problem that the header and the beater are difficult to be quickly adjusted in the prior art. By respectively establishing the corresponding relationship between the header height adjustment information and the beater speed adjustment information and the related parameters, the self-adaptive regulation of the header and the beater is realized. The header height and the beater speed can be accurately adjusted according to the actual operation conditions, such as the ground undulation, the crop density and the like. The accuracy and the adaptability of the regulation are improved. The header height and the beater speed are more accurately adjusted by comprehensively considering the factors such as the height of the two sides of the header and the ground, the operation speed, the crop density and the like, so that the harvesting operation is more smooth, and the problems such as the header damage and the crop omission are reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent agricultural machinery, and particularly relates to a self-adaptive control method for a harvesting machine header. BACKGROUND

[0002] The header is an important part of the harvesting machine, and the coordinated operation of each working component of the header directly affects the field working efficiency of the harvesting machine. During the operation of the grain harvesting machine, there are great differences in the terrain and soil conditions of different regions, and the meteorological conditions also cause differences in the ground flatness and crop harvesting state. Even in the same land block, due to the differences in planting conditions, the crop height and density are not uniform, which requires the operator to continuously adjust the height of the harvesting machine header, the speed of the reel, the position of the reel and other working states according to the crop state, the harvesting environment and the machine operation speed, otherwise, the situations such as blockage, missing harvesting, header touching the ground, machine overload or underload, etc. will occur, resulting in a decrease in working efficiency and an increase in failure rate.

[0003] A device and control method for adjusting the height of an automatic harvesting machine header are disclosed in Chinese Patent (Publication No. CN118489426A, Publication Date 20240816), which can adjust the height of the automatic harvesting machine header. The sensing device can sense the ground height in real time, and the control system can determine the control of the corresponding hydraulic cylinder stretching or the lifting of the overall harvesting device according to the ground height in front of the flexible header unit. The local header height or the overall header height can be adjusted according to the actual situation to ensure that the header cutter is not damaged. However, the harvesting machine header adjustment effect is poor, and the working state of the reel also affects the harvesting efficiency in addition to the header height and the header posture. The current control method is difficult to cooperatively adjust the header and the reel, resulting in low working efficiency and high failure rate. SUMMARY

[0004] The purpose of the present application is to provide a harvesting machine header self-adaptive control method to solve the problems existing in the prior art. By establishing the corresponding relationship between the header height adjustment information and the reel speed adjustment information and the related parameters respectively, the cooperative self-adaptive control of the header and the reel is realized. The header height and the reel speed can be accurately adjusted according to the actual working conditions such as ground undulation, crop density, etc. The accuracy and adaptability of the control are improved. The header height and the reel speed are more accurately adjusted by comprehensively considering factors such as the distance between the two sides of the header and the ground, the working speed, the crop density, etc. The harvesting operation is smoother through the cooperative control of the header and the reel, thereby improving the harvesting efficiency and reducing the failure rate.

[0005] To solve the above problems, the following solutions are adopted:

[0006] A harvesting machine header self-adaptive control method, comprising:

[0007] Obtaining the working speed of the header, height information of both sides of the header and images of crops in front of the header, processing the images to obtain crop density according to the image information of the crops in front of the header;

[0008] Determine corresponding height adjustment information of the header based on the height information of both sides of the header, the working speed and a preset profiling control model, and determine corresponding speed adjustment information of the windrower based on the crop density, the working speed and a preset windrower speed control model; wherein the preset profiling control model is a corresponding relationship between the height information of both sides of the header, the working speed and the height adjustment information of the header, and the preset windrower speed control model is a corresponding relationship between the crop density, the working speed and the speed adjustment information of the windrower;

[0009] Control the height of the header according to the height adjustment information of the header, and control the speed of the windrower according to the speed adjustment information of the windrower.

[0010] Further, the method further comprises:

[0011] Processing the images to obtain lodging state information, and determining corresponding position adjustment parameters of the windrower based on the lodging state information and a preset windrower position control model; the preset windrower position control model is a corresponding relationship between the lodging state information and the position adjustment parameters of the windrower;

[0012] Control the position of the windrower according to the position adjustment parameters of the windrower.

[0013] Further, the lodging state information includes forward lodging and reverse lodging relative to the working direction, and the position adjustment parameters of the windrower include lifting and lowering information of the windrower and forward and backward stretching information of the windrower.

[0014] Further, the control of the position of the windrower according to the position adjustment parameters of the windrower comprises:

[0015] When the lodging is forward, adjust the windrower to stretch forward to a front end position, and adjust the windrower to descend to a bottom end position;

[0016] When the lodging is reverse, adjust the windrower to shrink backward to a rear end position, and adjust the windrower to descend to a bottom end position.

[0017] Further, the method further comprises:

[0018] Obtain test data under different speeds and different crop densities through field tests, and establish a BP neural network model using the test data corresponding to the working parameters;

[0019] Optimize the BP neural network model using a genetic algorithm, and establish an optimal control parameter table of two-dimensional profiling control of the header under different working speeds and an optimal control parameter table of speed control of the windrower under different density conditions;

[0020] The preset profile control model and the preset reel speed control model are established according to optimal control parameter tables of longitudinal and transverse profile control of the header at different working speeds and optimal control parameter tables of the reel speed control under different density conditions.

[0021] Further, when the preset profile control model is established, the influencing factors of the longitudinal and transverse profile control of the header include the working speed, the PID control parameter, the header height target value, the header height measured value, and the deviation of the header height target value from the actual value.

[0022] Further, when the preset reel speed control model is established, the influencing factors of the reel speed control include the working speed, the crop density value, the PID control parameter, the reel speed measured value, and the reel speed to working speed ratio.

[0023] Further, the processing of the image to obtain the crop density comprises:

[0024] An image of crops in front of the header is acquired, the image is preprocessed, the image is gridded, and the color feature and the texture feature of the ear are extracted;

[0025] The extracted feature data is standardized, a classification learning method is used to classify the image patches in the image, and then the ear contour is recognized.

[0026] The ear contour image is refined through hole filling, the total number of the ears in the image is calculated, the number of the ears per unit area is obtained, and then the crop density is obtained.

[0027] Further, the working speed is acquired through a traveling speed sensor, the height information of the two sides of the header is acquired through a header longitudinal and transverse profiling device, the reel speed is acquired through a reel speed sensor, and the image is acquired through an industrial camera on the harvester.

[0028] Further, the height of the header is adjusted through a header lifting hydraulic cylinder, and the reel speed is adjusted through a hydraulic motor.

[0029] Compared with the prior art, the present application has the advantages and positive effects that:

[0030] In view of the problem that the header and the reel are difficult to be quickly adjusted in the prior art, by establishing the corresponding relationship between the header height adjustment information and the related parameters and the corresponding relationship between the reel speed adjustment information and the related parameters, the self-adaptive control of the header and the reel is realized, the header height and the reel speed can be accurately adjusted according to the actual working conditions, such as the ground undulation, the crop density, etc., the accuracy and the adaptability of the control are improved, the header height and the reel speed are more accurately adjusted by comprehensively considering the factors such as the height of the two sides of the header from the ground, the working speed, and the crop density, the harvesting operation is smoother, and the problems such as the damage of the header and the omission of crops are reduced. BRIEF DESCRIPTION OF DRAWINGS

[0031] The accompanying drawings, which form a part of this specification, are included to provide a further understanding of the application, and are incorporated herein by reference. The embodiments disclosed herein are illustrative of the application and are not meant to limit or restrict the scope of the application in any way.

[0032] Figure 1 The schematic diagram for the installation of the harvester header in the embodiment 1 of the application.

[0033] Figure 2 The schematic diagram for the longitudinal and transverse square device of the header in the embodiment 1 of the application Figure 1 .

[0034] Figure 3 The schematic diagram for the longitudinal and transverse square device of the header in the embodiment 1 of the application Figure 2 .

[0035] In the figure, 1, header body; 2, height detection device of the beater; 3, lifting hydraulic cylinder of the beater; 4, longitudinal and transverse profiling device of the header; 5, beater; 6, rotation speed sensor of the beater; 7, front and rear action hydraulic cylinder of the beater; 8, front and rear position detection device of the beater; 9, industrial camera; 10, profiling rod; 11, base; 12, connecting spherical hinge of the profiling rod; 13, hanging belt; 14, counterweight; 15, connecting bearing; 16, guard plate; 17, direction adjusting hole of the counterweight; 18, counterweight hanging device; 19, base hanging device; 20, angle detection element. DETAILED DESCRIPTION

[0036] Embodiment 1

[0037] In a typical embodiment of the application, as shown in Figures 1-3 , a self-adaptive control method for a harvester header is given.

[0038] A self-adaptive control method for a harvester header, comprising:

[0039] obtaining the working speed of the header, the height information of both sides of the header, and the image of the crops in front of the header, and processing the image to obtain the crop density;

[0040] determining the corresponding height adjustment information of the header based on the height information of both sides of the header, the working speed, and a preset profiling control model, and determining the corresponding rotation speed adjustment information of the beater based on the crop density, the working speed, and a preset rotation speed control model of the beater; wherein the preset profiling control model is the corresponding relationship between the height information of both sides of the header, the working speed, and the height adjustment information of the header, and the preset rotation speed control model of the beater is the corresponding relationship between the crop density, the working speed, and the rotation speed adjustment information of the beater;

[0041] controlling the height of the header according to the height adjustment information of the header, and controlling the rotation speed of the beater according to the rotation speed adjustment information of the beater.

[0042] The working speed is different, which affects the working state of the header and the reel; the height information of the two sides of the header and the ground is used to determine the height adjustment information of the header based on a preset profiling control model, and the image of the crops in front of the header is processed to obtain the crop density, so as to determine the reel speed adjustment information.

[0043] As shown in Figure 1 , the working speed is obtained by a traveling speed sensor, the height information of the two sides of the header is obtained by a header longitudinal and lateral profiling device, the image is obtained by an industrial camera on the harvester, and the reel speed is obtained by a reel speed sensor. The height of the header is adjusted by a header lifting hydraulic cylinder, and the reel speed is adjusted by a hydraulic motor.

[0044] Specifically, in combination with Figures 1-3 , the harvester includes an information acquisition device, a controller, and an execution mechanism. The information acquisition device includes a header longitudinal and lateral profiling device 4, a traveling speed sensor, a reel speed sensor 6, an industrial camera 9, and a reel position detection device. The working state of the harvester header is obtained by using the information acquisition device.

[0045] The industrial camera 8 is installed at the top of the front end of the harvester and takes real-time photos of the crops in front; the traveling speed sensor is installed on the walking system of the harvester; the reel speed sensor 6 is installed on the reel support at the end of the reel rotating shaft; the reel position detection device includes a reel height detection device 2 and a reel front and rear position detection device 8. One end of the reel height detection device 2 is installed on the header body 1, and the other end is connected with the reel support; one end of the reel front and rear position detection device 8 is installed on the reel support, and the other end is connected with the end of the reel rotating shaft.

[0046] The header longitudinal and lateral profiling device 4 is installed at the front end of the header, one on each left and right side, and includes a profiling rod 10, a base 11, a profiling rod connecting spherical hinge 12, a hanging belt 13, a counterweight 14, a connecting bearing 15, a guard plate 16, a counterweight direction adjusting hole 17, a counterweight hanging device 18, a base hanging device 19, and an angle detection element 20. The base 13 is installed at the front end of the header by bolts; the profiling rod 10 is connected with the rotating shaft through the spherical hinge 12; the rotating shaft is installed on the base through the bearing 15; the counterweight 5 is installed at the end of the profiling rod 10 by bolts, and the rotating angle of the counterweight 14 can be adjusted through the direction adjusting hole 17 on the profiling rod 10; the hanging belt 13 is hung on the base 11 through the hanging device 18 on the base and the counterweight.

[0047] The executive mechanism comprises: a cutting platform body 1, a reel lifting hydraulic cylinder 3, a reel 5, a reel lifting hydraulic cylinder 3, a reel front and rear action hydraulic cylinder 7, and a proportional valve group; the cutting platform body is connected with a bridge; one end of the reel front and rear action hydraulic cylinder 7 is fixed on a reel support, and the telescopic end is installed on the end of a reel rotating shaft; one end of the reel lifting action hydraulic cylinder 3 is fixed on the cutting platform body 1, and the telescopic end is installed on the reel support.

[0048] In the embodiment, the PID control parameters of the self-adaptive regulation and control method of the harvester cutting platform are based on the BP neural network model and the genetic algorithm optimization.

[0049] 1. Obtain test data under different speeds and different crop densities through field tests, and establish a BP neural network using a data set corresponding to the operation parameters.

[0050] (1) The influencing factors of the cutting platform longitudinal and transverse two-dimensional profiling regulation and control include operation speed, PID control parameters, cutting platform height target value, cutting platform height measurement value, cutting platform height target value and actual value deviation, etc. The data set of each influencing factor parameter under a certain operation speed is selected to establish a BP neural network. A three-layer network structure is adopted, and the input layer, hidden layer and output layer are respectively represented by i, j and l, and the superscripts are respectively represented by 1, 2 and 3. The input layer node is 4, the hidden layer node is 7, and the output node number of the output layer is 3. The input node and the output node are shown in Table 1.

[0051] Table 1

[0052]

[0053] The output of the neural network input layer is: i (1) x(i), (i=1, 2, 3, 4), the input layer variables x(i) are respectively the machine operation speed v, the cutting platform height target value H, the cutting platform height measurement value h, and the cutting platform height measurement value and the target value deviation Δe. The input and output of the neural network hidden layer are: wherein: NET j (2) is the input of the jth neuron of the hidden layer; W ji is the weight value between the ith neuron of the input layer and the jth neuron of the hidden layer; is the output of the ith neuron of the input layer; is the output of the jth neuron of the hidden layer. The input and output of the neural network output layer are respectively: wherein: NET l (3) is the input of the lth neuron of the output layer; W lj(k) is the weight value between the jth neuron of the hidden layer and the lth neuron of the output layer; is the output of the lth neuron of the output layer.

[0054] (2) The influencing factors of the reel speed regulation include the working speed, the crop density value, the PID control parameters, the reel speed measurement value, the reel speed and working speed ratio, etc. The data set of each influencing factor parameter under a certain crop density condition is selected to establish a BP neural network. A three-layer network structure is adopted, the input layer node is 4, the number of hidden layer nodes is 8, and the number of output nodes of the output layer is 3. The input nodes and output nodes are shown in Table 2.

[0055] Table 2

[0056]

[0057] The output of the input layer of the neural network is represented as: O i (1) = x(i), (i = 1, 2, 3, 4); the input and output of the hidden layer of the neural network are: The input and output of the output layer of the neural network are:

[0058] In the two BP neural network models constructed, the activation function of the hidden layer is a Tanh function The variable is mapped to [-1, 1]; the output layer parameters K p , K i , K D cannot be negative, and the activation function adopts a Sigmoid function The variable is mapped to [0, 1].

[0059] 2. The genetic algorithm is used to optimize the BP neural network model. First, all the weights and threshold values in the neural network are combined into individuals in the population; the error between the predicted output value and the actual output of the neural network is used to establish the fitness function of GA; the “survival of the fittest” is used to replace the back propagation process in the BP neural network; n generations are bred, and the individual with the highest fitness is selected from the population, which is the optimal weight and threshold value of the model; the optimal individual obtained is used as the initial solution of the BP algorithm, and the neural network is further optimized by using the BP algorithm; when the preset optimization target is reached or a sufficient number of iterations are performed, the optimal parameter combination under the working speed or crop density condition is output, and the values of the optimal control parameters K p , K i , K D are obtained.

[0060] The optimal control parameter table of the longitudinal and transverse two-dimensional profiling regulation of the header under different working speeds and the optimal control parameter table of the reel speed regulation under different density conditions are established by using the above method. The optimal control parameter table of the longitudinal and transverse two-dimensional profiling regulation of the header serves as a preset profiling regulation model, and can determine the corresponding header height adjustment information based on the height information of the two sides of the header and the working speed, so as to meet the demand of the header height adjustment. The optimal control parameter table of the reel speed regulation serves as a preset reel speed regulation model, and can determine the corresponding reel speed adjustment information based on the measured values of the crop density, the working speed and the reel speed, so as to meet the demand of the reel speed adjustment.

[0061] 3. Adjusting the longitudinal and transverse two-dimensional height of the header and the reel speed according to the optimal parameters

[0062] (1) Profiling regulation process of the longitudinal and transverse two-dimensional of the header

[0063] During the process of the machine operation, the two longitudinal and transverse profiling devices 4 on the two sides of the header detect the height fluctuation of the working ground in real time. When it is detected that the header height value on one side of the header deviates from the set value by Δe and does not meet the set deviation value range [Δe min Δe max ], the controller calls the optimal K p , K i , K D value under the speed, outputs the PWM signal to control the opening of the electromagnetic proportional valve, adjusts the action of the lifting hydraulic cylinder on the side, and realizes the rapid adjustment of the header height.

[0064] (2) Reel speed regulation process

[0065] Firstly, the number of grain ears in a unit area, i.e. the crop density ρ, is measured. In order to prevent the overlapping or missing of the shooting area of the industrial camera 4, the condition must be met, where T is the interval time of the camera shooting, D is the length of the camera field of view, and v is the machine forward speed. After the image acquisition, the shooting image is preprocessed by enhancing the image contrast, the image is gridded and the color features and texture features of the grain ear are extracted, the extracted feature data is standardized, the image patches in the image are classified by using the classification learning method, and then the grain ear contour is identified. The grain ear contour image is thinned by hole filling, the total number of grain ears in the image is calculated, and the number of grain ears in a unit area is obtained.

[0066] When the crop density ρ changes, the ratio coefficient k of the reel speed to the vehicle speed does not meet the set deviation value range [k min k max ], the controller calls the optimal K p , K i , K DThe value is output to control the opening of the electromagnetic proportional valve, to adjust the rotating speed of the reel, and to realize self-adaptive adjustment of the rotating speed of the reel and the crop density.

[0067] In addition, the image is processed to obtain the lodging state information, the corresponding reel position adjustment parameter is determined based on the lodging state information and a preset reel position control model, the preset reel position control model is a corresponding relationship between the lodging state information and the reel position adjustment parameter, and the reel position is controlled according to the reel position adjustment parameter.

[0068] Specifically, in the embodiment, the lodging state information of the crops is obtained by processing the image of the crops in front of the header. A preset reel position control model is established, and the model is a corresponding relationship between the lodging state information and the reel position adjustment parameter.

[0069] When the lodging state information includes forward lodging, the reel is adjusted to stretch forward to a front end position and to descend to a bottom end position according to the reel position adjustment parameter. When the lodging state information includes reverse lodging, the reel is adjusted to shrink backward to a rear end position and to descend to the bottom end position. In this way, the reel position is accurately adjusted according to the lodging state, the actual growth of the crops is better adapted, the harvesting efficiency is improved, and the loss of the crops is reduced.

[0070] Specifically, the control method of the reel in the lodging state of the crops can determine the lodging state of the crops by image recognition according to the image of the crops captured by the industrial camera 9. A BP neural network model in the lodging state of the crops can also be constructed according to the operation data, and a parameter table is set. The grains are analyzed to be in the lodging state according to the image of the grains. When the lodging state is forward lodging, the forward and backward action hydraulic cylinder 7 of the reel is adjusted to stretch to the limit state, so that the reel is in the front end operation position, and the lifting hydraulic cylinder 3 of the reel is lowered to the lowest position. When the lodging state is reverse lodging, the forward and backward action hydraulic cylinder 7 of the reel is adjusted to shrink to the limit state, so that the reel is in the rear end operation position, and the lifting hydraulic cylinder 3 of the reel is lowered to the lowest position. After the reel position is adjusted, the optimal control parameter is called to adjust the rotating speed of the reel according to the machine speed and the crop density value.

[0071] The above merely provides preferred embodiments of the present application but should not be used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for adaptive regulation of a harvester header, characterized in that, The method comprises the following steps: acquiring the working speed of the header, the height information of the two sides of the header, and the image of the crops in front of the header, processing the image to obtain the crop density according to the image information of the crops in front of the header; determining the corresponding height adjustment information of the header based on the height information of the two sides of the header, the working speed, and a preset profiling control model, and determining the corresponding rotation speed adjustment information of the windrower based on the crop density, the working speed, and a preset rotation speed control model of the windrower; wherein the preset profiling control model is the corresponding relationship between the height information of the two sides of the header, the working speed, and the height adjustment information of the header, and the preset rotation speed control model of the windrower is the corresponding relationship between the crop density, the working speed, and the rotation speed adjustment information of the windrower; controlling the height of the header according to the height adjustment information of the header, and controlling the rotation speed of the windrower according to the rotation speed adjustment information of the windrower. The method further comprises the following steps: acquiring the test data under different speeds and different crop densities through field tests, and establishing a BP neural network model by using the test data corresponding to the working parameters; optimizing the BP neural network model by using a genetic algorithm, and establishing an optimal control parameter table of the longitudinal and lateral profiling control of the header under different working speeds and an optimal control parameter table of the rotation speed control of the windrower under different density conditions; establishing the preset profiling control model and the preset rotation speed control model of the windrower according to the optimal control parameter table of the longitudinal and lateral profiling control of the header under different working speeds and the optimal control parameter table of the rotation speed control of the windrower under different density conditions.

2. The harvester header self-adaptive control method of claim 1, wherein, The method further comprises the following steps: processing the image to obtain the lodging state information, determining the corresponding position adjustment parameter of the windrower based on the lodging state information and a preset position control model of the windrower, and the preset position control model of the windrower being the corresponding relationship between the lodging state information and the position adjustment parameter of the windrower; controlling the position of the windrower according to the position adjustment parameter of the windrower.

3. The harvester header self-adaptive control method of claim 2, wherein, The lodging state information includes forward lodging and reverse lodging relative to the working direction, and the position adjustment parameter of the windrower includes lifting and lowering information and forward and backward stretching information of the windrower.

4. The harvester header self-adaptive control method of claim 3, wherein, The controlling of the position of the windrower according to the position adjustment parameter of the windrower comprises the following steps: when the lodging state is forward lodging, adjusting the windrower to stretch forward to a front end position and to lower to a bottom end position; when the lodging state is reverse lodging, adjusting the windrower to shrink backward to a rear end position and to lower to the bottom end position.

5. The harvester header self-adaptive control method of claim 1, wherein, When the preset profiling control model is established, the influencing factors of the longitudinal and lateral profiling control of the header include the working speed, PID control parameters, a target value of the height of the header, a measured value of the height of the header, and the deviation between the target value and the actual value of the height of the header.

6. The harvester header self-adaptive control method of claim 5, wherein, When the preset rotation speed control model of the windrower is established, the influencing factors of the rotation speed control of the windrower include the working speed, the crop density value, PID control parameters, a measured value of the rotation speed of the windrower, and the ratio of the rotation speed of the windrower to the working speed.

7. The harvester header self-adaptive control method of claim 1, wherein, The processing of the image to obtain the crop density comprises the following steps: acquiring the image of the crops in front of the header, pre-processing the photographed image, and gridizing the image and extracting the color features and texture features of the grains; standardizing the extracted feature data, classifying the image patches in the image by using a classification learning method, and then identifying the grain contours; refining the grain contour image by hole filling, calculating the total number of grains in the image, obtaining the number of grains per unit area, and then obtaining the crop density.

8. The harvester header self-adaptive control method of claim 1, wherein, The working speed is acquired by a traveling speed sensor, the height information on both sides of the header is acquired by a header longitudinal and lateral profiling device, the reel rotating speed is acquired by a reel rotating speed sensor, and the image is acquired by an industrial camera on the harvester.

9. The harvester header self-adaptive control method of claim 1 or 8, wherein, The height of the header is adjusted by a header lifting hydraulic cylinder drive, and the reel rotating speed is adjusted by a hydraulic motor.

Images