Self-adaptive mountain land advancing method for agricultural machine based on hydrostatic stepless speed change

Abstract

The invention discloses an agricultural machine self-adaptive mountainous region advancing method based on hydrostatic stepless speed change, and relates to the field of vehicle power. The method comprises the following steps: constructing an agricultural machine sensing system and a hydrostatic stepless speed change system; the pitch angle, the real-time acceleration and the engine rotating speed of the agricultural machine are obtained through an agricultural machine sensing system; through a recurrent neural network dynamics operation model, the accelerator opening degree and the valve opening degree are obtained through calculation; the engine is adjusted according to the accelerator opening, the hydrostatic stepless speed change system is adjusted according to the valve opening, and the agricultural machine is driven to advance in mountainous regions. The terrain and the motion state of the agricultural machine are obtained in real time, the smooth speed change capacity of the hydrostatic stepless speed change system is utilized, an operation method combining dynamics and a neural network is introduced, and the transmission ratio of engine power to the hydrostatic stepless speed change system is adjusted in a linkage mode, so that the agricultural machine stably advances and operates in a self-adaptive mode in a complex mountain land, and the working efficiency is improved. And the operation efficiency and the safety of the agricultural machine are greatly improved.

Description

Technical Field

[0001] The invention relates to the field of vehicle power, and in particular to an agricultural machinery adaptive mountain travel method based on hydrostatic continuously variable speed change. Background Art

[0002] Traditional agricultural machinery has stable performance when operating on flat terrain, but in complex terrain such as mountains, due to the undulating terrain, it is difficult for machinery to maintain a stable speed and power output, resulting in low operating efficiency and even possible safety accidents.

[0003] Hydrostatic Transmission (HST) is a transmission technology used in agricultural machinery, construction equipment, industrial machinery and other fields. It can provide smooth acceleration and deceleration, reduce mechanical wear and extend service life. HST allows precise control of speed and torque, adapts to different operating requirements, improves operating efficiency and quality, and has great advantages in driving agricultural machinery to travel in mountainous areas. However, the existing technology lacks an adaptive control method with fast response speed and precise power control, and cannot make agricultural machinery in operation travel smoothly and efficiently in mountainous areas. Summary of the invention

[0004] In response to the above-mentioned deficiencies in the prior art, the present invention provides an agricultural machinery adaptive mountain travel method based on hydrostatic continuously variable speed change, which solves the problem that the drive control of existing agricultural machinery cannot enable the agricultural machinery in operation to move smoothly, efficiently and safely in complex terrains such as mountains.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0006] A method for agricultural machinery to adaptively travel in mountainous terrain based on hydrostatic continuously variable speed change comprises the following steps:

[0007] S1, construct agricultural machinery sensing system and hydrostatic continuously variable transmission system;

[0008] S2, obtaining the pitch angle, real-time acceleration and engine speed of the agricultural machinery through the agricultural machinery sensing system;

[0009] S3, according to the pitch angle, real-time acceleration and engine speed, the throttle opening and valve opening are calculated by using a recursive neural network dynamics calculation model;

[0010] S4, adjusting the engine according to the throttle opening, and adjusting the hydrostatic continuously variable transmission system according to the valve opening, so as to drive the agricultural machine to move in the mountains.

[0011] The beneficial effects of the present invention are as follows: the present invention obtains the terrain and movement status of the agricultural machinery in real time through the agricultural machinery sensing system, utilizes the smooth speed change capability of the hydrostatic continuously variable transmission system, introduces a calculation method combining dynamics and neural networks, and jointly adjusts the engine power and the transmission ratio of the hydrostatic continuously variable transmission system, so that the agricultural machinery can adaptively and smoothly move and operate stably in complex mountainous areas, thereby greatly improving the operating efficiency and safety of the agricultural machinery.

[0012] Furthermore, the agricultural machinery sensing system comprises:

[0013] Pitch angle sensor, used to detect the pitch angle of agricultural machinery;

[0014] Acceleration sensor, used to detect the real-time acceleration of agricultural machinery;

[0015] Engine speed sensor, used to detect the engine speed of agricultural machinery.

[0016] Furthermore, the hydrostatic continuously variable transmission system comprises: a hydraulic pump, an oil circuit valve and a hydraulic motor;

[0017] The hydraulic pump is used to convert the mechanical energy of the agricultural machinery engine into the hydraulic energy of the oil circuit, and its rotor end is connected to the agricultural machinery engine, and its pressure port is connected to one end of the oil circuit valve through the oil circuit;

[0018] The oil circuit valve is used to adjust the flow and pressure of the oil circuit through the valve opening to smoothly change the transmission ratio of the hydrostatic continuously variable transmission system, and the other end of the oil circuit valve is connected to the hydraulic motor through the oil circuit;

[0019] The hydraulic motor is used to convert hydraulic energy back into mechanical energy, and its rotor end is connected to the transmission shaft of the agricultural machinery to drive the agricultural machinery to move.

[0020] The beneficial effect of the above further scheme is that the hydrostatic continuously variable transmission system of the present invention adjusts the oil flow and pressure of the hydraulic pump through the hydraulic motor through the oil circuit valve, so that the transmission ratio can change smoothly with the change of the valve opening. There is no sense of frustration of manual transmission and dual clutch transmission.

[0021] Furthermore, the recursive neural network dynamics operation model includes:

[0022] The acceleration difference calculation module is used to calculate the acceleration difference according to the target acceleration and the real-time acceleration;

[0023] The dynamic calculation module is used to calculate the torque requirement based on the pitch angle and acceleration difference;

[0024] The recursive neural network operation module is used to calculate the throttle opening and the valve opening according to the torque demand and the engine speed.

[0025] The beneficial effect of the above further solution is that when the agricultural machinery is traveling on complex terrain such as mountains, in order to steadily accelerate or travel at a constant speed, the torque needs to be adjusted in real time according to the slope, that is, the pitch angle of the agricultural machinery. The above solution can control the engine throttle and the transmission ratio of the hydrostatic continuously variable transmission system in a linked manner to achieve the smooth travel of the agricultural machinery at the most suitable engine speed.

[0026] Furthermore, the expression of the acceleration difference calculation module is:

[0027] ,

[0028] Among them, a e is the acceleration difference, a is the real-time acceleration, a goal is the target acceleration;

[0029] When the agricultural machine is started, the target speed and the target acceleration are set by the user; after the agricultural machine reaches the target speed, the target acceleration is adjusted to 0.

[0030] Furthermore, the expression of the dynamics calculation module is:

[0031] ,

[0032] Where M is the torque demand, a e is the acceleration difference, g is the gravitational acceleration, μ is the tire rolling friction coefficient, cos(∙) is the cosine function, θ is the pitch angle, m is the vehicle mass, r is the tire radius, h is the vehicle height, and sin(∙) is the sine function.

[0033] The beneficial effect of the above further scheme is: the above formula is based on dynamics, and calculates the torque required by the agricultural machinery according to the acceleration demand of the agricultural machinery, that is, the acceleration difference and the slope of the terrain on which the agricultural machinery is located, that is, the pitch angle of the agricultural machinery, so that the recursive neural network operation module can perform engine throttle and hydrostatic continuously variable transmission system transmission ratio.

[0034] Furthermore, the recursive neural network operation module includes an update layer, a memory layer, and an output layer, and the output layer expression is:

[0035] ,

[0036] ,

[0037] Among them, L oil (t) is the throttle opening at time t, γ(t) is the update layer output value at time t, and w 1,oil,M is the output layer throttle torque weight coefficient, M(t) is the torque demand at time t, and w 1,oil,c is the throttle speed weight coefficient of the output layer, c(t) is the engine speed at time t, is the n1th recursive coefficient of the output layer throttle, h(t-n1) is the output value of the memory layer at time t-n1, b 1,oil is the throttle bias coefficient of the output layer, N1 is the recursive operation length of the output layer, L gate (t) is the valve opening at time t, w 1,gate,M is the output layer valve-torque weight coefficient, w 1,gate,c is the output layer valve-speed weight coefficient, is the n1th recursive coefficient of the output layer valve, b 1,gate is the output layer valve bias coefficient.

[0038] Furthermore, the update layer expression of the recursive neural network operation module is:

[0039] ,

[0040] Among them, γ(t) is the output value of the update layer at time t, w 2,M is the torque weight coefficient of the update layer, M(t) is the torque demand at time t, and w 2,c is the update layer speed weight coefficient, c(t) is the engine speed at time t, is the n2th recursive coefficient of the update layer, γ(t-n2) is the output value of the update layer at time t-n2, b2 is the bias coefficient of the update layer, and N2 is the recursive operation length of the update layer.

[0041] Furthermore, the memory layer expression of the recursive neural network operation module is:

[0042] ,

[0043] Among them, h(t) is the output value of the memory layer at time t, w 3,M is the torque weight coefficient of the memory layer, M(t) is the torque demand at time t, and w 3,c is the speed weight coefficient of the memory layer, c(t) is the engine speed at time t, is the n3th recursive coefficient of the memory layer, h(t-n3) is the output value of the memory layer at time t-n3, b3 is the bias coefficient of the memory layer, and N3 is the recursive operation length of the memory layer.

[0044] The beneficial effect of the above further scheme is that the influence of engine power and transmission ratio of the hydrostatic continuously variable transmission system on the torque and engine speed of the agricultural machinery is accumulated in the time domain. The present invention sets up a special neural network based on time domain recursive operation, and adjusts the power and transmission ratio in a linked manner through the engine throttle opening and the hydrostatic continuously variable transmission system oil circuit valve opening, so that the engine speed of the agricultural machinery reaches an appropriate state and obtains a torque that allows the agricultural machinery to move steadily. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flow chart of a method for adaptive mountaineering of agricultural machinery based on hydrostatic continuously variable speed change provided by an embodiment of the present invention;

[0046] Figure 2 A structural diagram of an agricultural machinery sensing system and a hydrostatic continuously variable transmission system according to an embodiment of the present invention;

[0047] Figure 3 It is a structural diagram of the recursive neural network dynamics operation model of an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0049] like Figure 1 As shown, in one embodiment of the present invention, a method for agricultural machinery to adaptively travel in mountainous terrain based on hydrostatic continuously variable speed transmission comprises the following steps:

[0050] S1. Build agricultural machinery sensing system and hydrostatic continuously variable transmission system.

[0051] like Figure 2 As shown, the agricultural machinery sensing system includes:

[0052] Pitch angle sensor, used to detect the pitch angle of agricultural machinery;

[0053] Acceleration sensor, used to detect the real-time acceleration of agricultural machinery;

[0054] Engine speed sensor, used to detect the engine speed of agricultural machinery.

[0055] In this embodiment, the pitch angle sensor and the acceleration sensor use MEMS inertial sensors, and the engine speed sensor uses a magnetoelectric sensor.

[0056] The hydrostatic continuously variable transmission system includes: a hydraulic pump, oil circuit valves and a hydraulic motor.

[0057] The hydraulic pump is used to convert the mechanical energy of the agricultural machinery engine into the hydraulic energy of the oil circuit. Its rotor end is connected to the agricultural machinery engine, and its pressure port is connected to one end of the oil circuit valve through the oil circuit; the oil circuit valve is used to adjust the flow and pressure of the oil circuit through the valve opening to smoothly change the transmission ratio of the hydrostatic continuously variable transmission system, and its other end is connected to the hydraulic motor through the oil circuit; the hydraulic motor is used to convert hydraulic energy back into mechanical energy, and its rotor end is connected to the drive shaft of the agricultural machinery to drive the agricultural machinery to move.

[0058] The hydrostatic continuously variable transmission system of the present invention adjusts the oil flow and pressure of the hydraulic pump through the hydraulic motor through the oil circuit valve, so that the transmission ratio can change smoothly with the change of the valve opening, without the frustration of manual transmission and dual clutch transmission.

[0059] S2. Obtain the pitch angle, real-time acceleration and engine speed of the agricultural machinery through the agricultural machinery sensing system.

[0060] S3. According to the pitch angle, real-time acceleration and engine speed, the throttle opening and valve opening are calculated through a recursive neural network dynamics calculation model.

[0061] S4, adjusting the engine according to the throttle opening, and adjusting the hydrostatic continuously variable transmission system according to the valve opening, so as to drive the agricultural machine to move in the mountains.

[0062] like Figure 3 As shown, the recurrent neural network dynamics operation model includes:

[0063] The acceleration difference calculation module is used to calculate the acceleration difference according to the target acceleration and the real-time acceleration. The expression is:

[0064] ,

[0065] Among them, a e is the acceleration difference, a is the real-time acceleration, a goal is the target acceleration.

[0066] In this embodiment, when the agricultural machine is started, the target speed and the target acceleration are set by the user; after the agricultural machine reaches the target speed, the target acceleration is adjusted to 0.

[0067] When agricultural machinery is traveling on complex terrain such as mountains, in order to steadily accelerate or travel at a constant speed, it is necessary to adjust the torque in real time according to the slope, that is, the pitch angle of the agricultural machinery. The dynamic calculation module provided by the present invention can control the engine throttle and the transmission ratio of the hydrostatic continuously variable transmission system in a linked manner to achieve the smooth travel of the agricultural machinery at the most suitable engine speed.

[0068] The dynamic calculation module is used to calculate the torque demand based on the pitch angle and acceleration difference. The expression is:

[0069] ,

[0070] Where M is the torque demand, a e is the acceleration difference, g is the gravitational acceleration, μ is the tire rolling friction coefficient, cos(∙) is the cosine function, θ is the pitch angle, m is the vehicle mass, r is the tire radius, h is the vehicle height, and sin(∙) is the sine function.

[0071] The above formula is based on dynamics. It calculates the torque required by the agricultural machinery according to the acceleration demand of the agricultural machinery, that is, the acceleration difference, and the slope of the terrain where the agricultural machinery is located, that is, the pitch angle of the agricultural machinery, so that the recursive neural network operation module can perform engine throttle and hydrostatic continuously variable transmission system transmission ratio.

[0072] The recursive neural network operation module is used to calculate the throttle opening and valve opening according to the torque demand and engine speed. It includes an update layer, a memory layer, and an output layer.

[0073] The output layer expression of the recursive neural network operation module is:

[0074] ,

[0075] ,

[0076] Among them, L oil (t) is the throttle opening at time t, γ(t) is the update layer output value at time t, and w 1,oil,M is the output layer throttle torque weight coefficient, M(t) is the torque demand at time t, and w 1,oil,c is the throttle speed weight coefficient of the output layer, c(t) is the engine speed at time t, is the n1th recursive coefficient of the output layer throttle, h(t-n1) is the output value of the memory layer at time t-n1, b 1,oil is the throttle bias coefficient of the output layer, N1 is the recursive operation length of the output layer, L gate (t) is the valve opening at time t, w 1,gate,M is the output layer valve-torque weight coefficient, w 1,gate,c is the output layer valve-speed weight coefficient, is the n1th recursive coefficient of the output layer valve, b 1,gate is the output layer valve bias coefficient.

[0077] The update layer expression of the recurrent neural network operation module is:

[0078] ,

[0079] Among them, γ(t) is the output value of the update layer at time t, w 2,Mis the torque weight coefficient of the update layer, M(t) is the torque demand at time t, and w 2,c is the update layer speed weight coefficient, c(t) is the engine speed at time t, is the n2th recursive coefficient of the update layer, γ(t-n2) is the output value of the update layer at time t-n2, b2 is the bias coefficient of the update layer, and N2 is the recursive operation length of the update layer.

[0080] The memory layer expression of the recursive neural network operation module is:

[0081] ,

[0082] Among them, h(t) is the output value of the memory layer at time t, w 3,M is the torque weight coefficient of the memory layer, M(t) is the torque demand at time t, and w 3,c is the speed weight coefficient of the memory layer, c(t) is the engine speed at time t, is the n3th recursive coefficient of the memory layer, h(t-n3) is the output value of the memory layer at time t-n3, b3 is the bias coefficient of the memory layer, and N3 is the recursive operation length of the memory layer.

[0083] The influence of engine power and transmission ratio of hydrostatic continuously variable transmission system on torque and engine speed of agricultural machinery is cumulative in time domain. The present invention sets up a special neural network based on time domain recursive operation, and adjusts power and transmission ratio in linkage through engine throttle opening and hydrostatic continuously variable transmission system oil circuit valve opening, so that the engine speed of agricultural machinery reaches a suitable state and obtains torque that allows the agricultural machinery to move steadily.

[0084] The recursive operation length of each layer of the recursive neural network operation module can be adjusted as needed. The larger the value, the stronger the model fitting ability, but the larger the amount of calculation. In this embodiment, it is set to 5. The weight coefficient, recursive coefficient and bias coefficient need to be trained by machine learning. Because the throttle opening acts on the engine and the valve opening acts on the hydrostatic continuously variable transmission system, it will cause changes in real-time acceleration and engine speed, forming a feedback loop. Therefore, it is not suitable to use an ordinary gradient descent algorithm for parameter training. In this embodiment, the existing genetic ant colony hybrid algorithm is used to train the above parameters. It is worth noting that in order to save fuel, this embodiment sets the boundary condition that the engine speed is less than or equal to 3000 revolutions per minute, and takes the faster the acceleration difference converges to 0 as the training goal. The training process will not be repeated.

[0085] In summary, the present invention obtains the terrain and movement status of the agricultural machinery in real time through the agricultural machinery sensing system, utilizes the smooth speed shifting ability of the hydrostatic continuously variable transmission system, introduces a calculation method that combines dynamics and neural networks, and jointly adjusts the engine power and the transmission ratio of the hydrostatic continuously variable transmission system, so that the agricultural machinery can adaptively move smoothly and operate stably in complex mountainous areas, which greatly improves the operating efficiency and safety of the agricultural machinery.

[0086] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0087] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A method for agricultural machinery to adaptively travel in mountainous terrain based on hydrostatic continuously variable speed, characterized in that: The following steps are involved: S1, construct agricultural machinery sensing system and hydrostatic continuously variable transmission system; S2, obtaining the pitch angle, real-time acceleration and engine speed of the agricultural machinery through the agricultural machinery sensing system; S3, according to the pitch angle, real-time acceleration and engine speed, the throttle opening and valve opening are calculated by using a recursive neural network dynamics calculation model; S4, adjusting the engine according to the throttle opening, and adjusting the hydrostatic continuously variable transmission system according to the valve opening, so as to drive the agricultural machine to move in the mountains.

2. The method for agricultural machinery to adaptively travel in mountainous terrain based on hydrostatic continuously variable speed according to claim 1 is characterized in that: The agricultural machinery sensing system comprises: Pitch angle sensor, used to detect the pitch angle of agricultural machinery; Acceleration sensor, used to detect the real-time acceleration of agricultural machinery; Engine speed sensor, used to detect the engine speed of agricultural machinery.

3. The method for agricultural machinery to self-adapt to mountain terrain based on hydrostatic continuously variable speed according to claim 1, characterized in that: The hydrostatic continuously variable transmission system comprises: a hydraulic pump, an oil circuit valve and a hydraulic motor; The hydraulic pump is used to convert the mechanical energy of the agricultural machinery engine into the hydraulic energy of the oil circuit, and its rotor end is connected to the agricultural machinery engine, and its pressure port is connected to one end of the oil circuit valve through the oil circuit; The oil circuit valve is used to adjust the flow and pressure of the oil circuit through the valve opening to smoothly change the transmission ratio of the hydrostatic continuously variable transmission system, and the other end of the oil circuit valve is connected to the hydraulic motor through the oil circuit; The hydraulic motor is used to convert hydraulic energy back into mechanical energy, and its rotor end is connected to the transmission shaft of the agricultural machinery to drive the agricultural machinery to move.

4. The method for agricultural machinery to self-adapt to mountain terrain based on hydrostatic continuously variable speed according to claim 1, characterized in that: The recursive neural network dynamics operation model includes: The acceleration difference calculation module is used to calculate the acceleration difference according to the target acceleration and the real-time acceleration; The dynamic calculation module is used to calculate the torque requirement based on the pitch angle and acceleration difference; The recursive neural network operation module is used to calculate the throttle opening and the valve opening according to the torque demand and the engine speed.

5. The method for agricultural machinery to self-adapt to mountain terrain based on hydrostatic continuously variable speed change according to claim 4, characterized in that: The expression of the acceleration difference calculation module is: ， Among them, a e is the acceleration difference, a is the real-time acceleration, a goal is the target acceleration; When the agricultural machine is started, the target speed and the target acceleration are set by the user; after the agricultural machine reaches the target speed, the target acceleration is adjusted to 0.

6. The method for agricultural machinery to self-adapt to mountain terrain based on hydrostatic continuously variable speed according to claim 4, characterized in that: The expression of the dynamics calculation module is: ， Where M is the torque demand, a e is the acceleration difference, g is the gravitational acceleration, μ is the tire rolling friction coefficient, cos(∙) is the cosine function, θ is the pitch angle, m is the vehicle mass, r is the tire radius, h is the vehicle height, and sin(∙) is the sine function.

7. The method for agricultural machinery to self-adapt to mountain terrain based on hydrostatic continuously variable speed change according to claim 4, characterized in that: The recursive neural network operation module includes an update layer, a memory layer, and an output layer, and the output layer expression is: ， ， Among them, L oil (t) is the throttle opening at time t, γ(t) is the update layer output value at time t, and w 1,oil,M is the output layer throttle torque weight coefficient, M(t) is the torque demand at time t, and w 1,oil,c is the throttle speed weight coefficient of the output layer, c(t) is the engine speed at time t, is the n1th recursive coefficient of the output layer throttle, h(t-n1) is the output value of the memory layer at time t-n1, b 1,oil is the throttle bias coefficient of the output layer, N1 is the recursive operation length of the output layer, L gate (t) is the valve opening at time t, w 1,gate,M is the output layer valve-torque weight coefficient, w 1,gate,c is the output layer valve-speed weight coefficient, is the n1th recursive coefficient of the output layer valve, b 1,gate is the output layer valve bias coefficient.

8. The method for agricultural machinery to self-adapt to mountain terrain based on hydrostatic continuously variable transmission according to claim 7, characterized in that: The update layer expression of the recursive neural network operation module is: ， Among them, γ(t) is the output value of the update layer at time t, w 2,M is the torque weight coefficient of the update layer, M(t) is the torque demand at time t, and w 2,c is the update layer speed weight coefficient, c(t) is the engine speed at time t, is the n2th recursive coefficient of the update layer, γ(t-n2) is the output value of the update layer at time t-n2, b2 is the bias coefficient of the update layer, and N2 is the recursive operation length of the update layer.

9. The method for agricultural machinery to self-adapt to mountain terrain based on hydrostatic continuously variable speed change according to claim 8, characterized in that: The memory layer expression of the recursive neural network operation module is: ， Among them, h(t) is the output value of the memory layer at time t, w 3,M is the torque weight coefficient of the memory layer, M(t) is the torque demand at time t, and w 3,c is the speed weight coefficient of the memory layer, c(t) is the engine speed at time t, is the n3th recursive coefficient of the memory layer, h(t-n3) is the output value of the memory layer at time t-n3, b3 is the bias coefficient of the memory layer, and N3 is the recursive operation length of the memory layer.

Images