A tractor dual-channel power decoupling energy management method and system based on load prediction

Abstract

The present application belongs to the field of energy management of agricultural equipment hybrid tractor, and particularly relates to a kind of energy management method and system of tractor double-channel power decoupling based on load prediction, comprising the following steps: collecting the operating state parameters of tractor, work load parameters and power system state parameters, construct the state vector of current time;Based on the load sequence in the preset history window and the state vector, the walking load prediction sequence and the PTO load prediction sequence in the prediction time domain are output through the load prediction model;Walking power channel power demand and PTO power channel power demand are calculated respectively, and the power tracking error and dynamic priority constraint of the two channels are established;The method can reduce the influence of PTO load mutation on walking speed, and reduce the influence of walking load mutation on PTO speed, improve the stability and fuel economy of tractor operation.

Description

Technical Field

[0001] This invention belongs to the field of energy management technology for hybrid tractors in agricultural equipment, specifically relating to an energy management method and system for dual-channel power decoupling of tractors based on load prediction. Background Technology

[0002] When tractors operate in the field, they typically have simultaneous needs for both travel power and PTO (Power Take-Off) power. The travel power channel meets the needs of vehicle movement, traction, and turning at the end of the field, while the PTO power channel drives implements such as rotary tillers, seeders, and balers. Compared to road vehicles, tractor PTO and traction loads are characterized by strong fluctuations, instability, cyclical impacts, and strong correlation with the operating environment. Existing energy management methods for hybrid tractors mostly employ rule-based control, instantaneous optimization, or equivalent fuel minimization strategies. These methods typically passively allocate power based on the current load, making it difficult to anticipate sudden changes in soil resistance, tillage depth, implement stalling, or PTO engagement shocks. Furthermore, travel power and PTO power often share the same engine or main drivetrain. A sudden increase in PTO load can easily cause a drop in engine speed and lead to vehicle speed fluctuations; conversely, a sudden increase in travel load can easily cause a drop in PTO speed, affecting work quality.

[0003] Therefore, an energy management method is needed that can predict future loads and calculate and coordinate the travel path and PTO path separately, so that the engine, motor, energy storage system and PTO actuator can coordinate control in advance in the prediction time domain, thereby improving fuel economy, dynamic response and operational stability. Summary of the Invention

[0004] The purpose of this invention is to provide an energy management method and system for dual-channel power decoupling of tractors based on load prediction. This method obtains the current operating state, working load, and power system state of the tractor through state perception, and uses a load prediction model to obtain the travel load and PTO load in the future prediction time domain. Power demand, power tracking error, and dynamic priority constraints are established in the travel power channel and the PTO power channel, respectively. The power distribution control quantity is solved under the constraints of the engine, motor, power battery, and PTO actuator through the energy management optimization module, and the control command is sent to each actuator, thereby reducing the dynamic coupling effect between the two power channels.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: An energy management method for dual-channel power decoupling of a tractor based on load prediction includes the following steps: collecting the tractor's operating status parameters, work load parameters, and power system status parameters to construct a state vector at the current moment; based on the load sequence within a preset historical window and the state vector, outputting a travel load prediction sequence and a PTO load prediction sequence in the prediction time domain through a load prediction model; calculating the power demand of the travel power channel based on the travel load prediction sequence, and calculating the power demand of the PTO power channel based on the PTO load prediction sequence, so that the travel power channel and the PTO power channel each form independent power demands; establishing the power tracking error of the travel power channel and the power tracking error of the PTO power channel in the prediction time domain, and calculating the power tracking error of the PTO power channel based on the PTO speed deviation, vehicle speed deviation, and power supply... The battery's SOC (State of Charge) status determines the dynamic priority weights of the driving power channel and the PTO (Power Toll Collection) power channel. When the PTO speed deviation exceeds a preset PTO speed deviation threshold, the priority of the PTO power channel is increased, and the power change rate of the driving power channel is limited. When the driving speed deviation exceeds a preset vehicle speed deviation threshold and the PTO speed is within the allowable range, the priority of the driving power channel is increased, and the driving power gap is compensated by the drive motor and the power battery. Using fuel consumption, power battery state of charge deviation, and engine dynamic changes as optimization objectives, and constrained by the output capabilities of the engine, motor, power battery, and PTO, the power distribution control quantity in the prediction time domain is solved. Based on the power distribution control quantity, control commands are output to the engine controller, drive motor controller, PTO actuator, and battery management system.

[0006] The operating status parameters include vehicle speed, acceleration, throttle opening, gear, engine speed, and PTO speed; the working load parameters include at least one of traction resistance, PTO torque, tillage depth, implement lifting status, soil moisture content, or working position; the power system status parameters include power battery SOC, battery allowable charge and discharge power, motor speed, motor torque, and current engine operating point.

[0007] The state vector is constructed as follows:

[0008] in For vehicle speed, Acceleration, Engine speed, For PTO rotation speed, For traction resistance, For PTO torque, This refers to the depth of tillage or working.

[0009] The load prediction model employs Long Short-Term Memory (LSTM) networks, gated recurrent units (GRUs), temporal convolutional networks, Transformer networks, or combinations thereof. Input, Output Where N is the historical window length and H is the prediction time domain length. The model is designed to predict traction resistance and PTO torque sequences over a time span. It is trained under supervised supervision using historical window state sequences and future load sequences collected under various operating conditions, including rotary tillage, plowing, sowing, and transportation. The loss function includes at least the walking load prediction error and the PTO load prediction error.

[0010] The power requirement of the walking power channel is as follows calculate, It is determined at least by rolling resistance, gradient resistance, acceleration resistance, air resistance, and predicted traction operation resistance. For walking transmission efficiency.

[0011] The power requirement of the PTO power channel is as follows calculate, To predict PTO torque, The target speed or measured speed for PTO. For PTO transmission efficiency.

[0012] The dual-channel power decoupling satisfies the requirement that the power demand of the walking power channel and the power demand of the PTO power channel are calculated and constrained separately. Independent power tracking error, power limit, power change rate limit and priority constraint are set for the walking power channel and the PTO power channel respectively, so as to reduce the impact of PTO load change on walking speed and the impact of walking load change on PTO speed.

[0013] Furthermore, the power tracking error of the driving power channel is determined by the deviation between the required power and the actual power of the driving power channel, while the power tracking error of the PTO power channel is determined by the deviation between the required power and the actual power of the PTO power channel. The penalty coefficients for the two power tracking errors in the optimization objective function are adjusted according to dynamic priority weights. The optimization objective function, with fuel consumption, battery state of charge deviation, engine dynamic changes, driving power channel power tracking error, and PTO power channel power tracking error as optimization objectives, is as follows:

[0014] in For fuel consumption rate, For the target SOC, This represents the change in engine speed. This represents the change in engine power. For the power tracking error of the walking power channel, For PTO power tracking error , , , , These are the weighting coefficients.

[0015] The dynamic priority weights satisfy the following: when the PTO speed deviation is greater than the preset PTO speed deviation threshold, the weight of the PTO power channel is increased; when the vehicle speed deviation is greater than the preset vehicle speed deviation threshold and the PTO speed is within the allowable range, the weight of the driving power channel is increased; when the power battery SOC is lower than the preset SOC lower limit protection value, the power battery discharge compensation power is limited.

[0016] Furthermore, the optimization objectives are fuel consumption, power battery state of charge deviation, engine dynamic changes, power tracking error of the driving power channel, and power tracking error of the PTO power channel. An optimization problem is constructed with the output capabilities of the engine, motor, power battery, and PTO as constraints. The constraints include engine speed and torque constraints, motor speed and torque constraints, power battery SOC upper and lower limit constraints, power battery charging and discharging power constraints, PTO speed allowable deviation constraints, driving power channel power change rate constraints, PTO power channel power change rate constraints, and power balance constraints, where the power battery power is positive for discharging and negative for charging.

[0017] Furthermore, the present invention also provides a tractor dual-channel power decoupling energy management system based on load prediction, comprising: a state perception module, a load prediction module, a dual-channel power decoupling module, an energy management optimization module, and an execution control module; the state perception module is used to collect tractor operating state parameters, work load parameters, and power system state parameters; the load prediction module is used to output a travel load prediction sequence and a PTO load prediction sequence in the prediction time domain; the dual-channel power decoupling module is used to calculate the power demand of the travel power channel and the power demand of the PTO power channel respectively; the energy management optimization module is used to solve the power distribution control quantities of the engine, motor, PTO, and power battery under constraints; and the execution control module is used to send the power distribution control quantities to the engine controller, drive motor controller, PTO actuator, and battery management system.

[0018] The PTO actuator includes at least one of a PTO motor, a PTO clutch, a PTO brake, or a PTO power converter, and the execution control module can perform closed-loop control of the PTO output torque or PTO output speed according to the PTO channel power requirements.

[0019] The beneficial effects of this invention are as follows: The energy management method and system for tractor dual-channel power decoupling based on load prediction provided by this invention have the following characteristics in practical use: 1. This invention can obtain future load change trends in advance through a load prediction model, and can complete the pre-allocation of engine and motor power before load changes, reducing response lag; by independently modeling the driving power channel and PTO power channel, and combining power tracking error and dynamic priority weight, it can reduce dynamic coupling interference between the two channels and improve vehicle speed stability and PTO speed stability. 2. This invention optimizes the time domain to ensure the engine operates in its most efficient region and reduces significant fluctuations in engine speed and power, thereby reducing fuel consumption. Through SOC constraints, battery power constraints, and SOC lower limit protection, the power battery participates in peak shaving and valley filling within permissible limits, extending the lifespan of the energy storage system. 3. The present invention is applicable to series, parallel, and hybrid tractors, and can also be extended to non-road agricultural machinery with PTO output. Attached Figure Description

[0020] Figure 1 This is a structural diagram of the dual-channel power decoupling energy management system for tractors based on load prediction, as described in this invention. Figure 2 This is a schematic diagram of the dual-channel power decoupling of the present invention; Figure 3 This is a flowchart of the energy management method based on load prediction according to the present invention; Figure 4 This is a table showing the component modules of the tractor dual-channel power decoupling energy management system based on load prediction, as described in this invention. Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It should be noted that, unless otherwise specified, all embodiments and preferred methods mentioned herein can be combined to form new technical solutions. Unless otherwise specified, all technical features and preferred features mentioned herein can be combined to form new technical solutions. Unless otherwise stated, the professional and scientific terms used herein have the same meaning as those familiar with the art. Furthermore, any methods or materials similar to or equivalent to the described content can also be applied to the present invention.

[0022] Specific Implementation Example 1: As per the appendix to the instruction manual Figure 1As shown, the purpose of this invention is to provide an energy management method and system for dual-channel power decoupling of a tractor based on load prediction. The dual-channel power decoupling energy management system includes: a state perception module, a load prediction module, a dual-channel power decoupling module, an energy management optimization module, and an execution control module. The state perception module is used to collect tractor operating state parameters, work load parameters, and power system state parameters. The load prediction module is used to output a prediction sequence of travel load and a prediction sequence of PTO load in the prediction time domain. The dual-channel power decoupling module is used to calculate the power demand of the travel power channel and the power demand of the PTO power channel respectively, and to establish power tracking error, power limiting, and dynamic priority constraints for the two channels. The energy management optimization module is used to solve for the power distribution control quantities of the engine, motor, PTO, and power battery under constraints. The execution control module is used to send the power distribution control quantities to the engine controller, drive motor controller, PTO actuator, and battery management system. The PTO actuator includes at least one of a PTO motor, a PTO clutch, a PTO brake, or a PTO power converter. The execution control module can perform closed-loop control of the PTO output torque or PTO output speed according to the PTO channel power demand. The energy management method for dual-channel power decoupling of a tractor includes the following steps: collecting the tractor's operating status parameters, work load parameters, and power system status parameters to construct a state vector at the current moment; based on the load sequence within a preset historical window and the state vector, outputting a travel load prediction sequence and a PTO load prediction sequence in the prediction time domain through a load prediction model; calculating the power demand of the travel power channel based on the travel load prediction sequence, and calculating the power demand of the PTO power channel based on the PTO load prediction sequence, so that the travel power channel and the PTO power channel each form independent power demands; establishing travel power channels separately in the prediction time domain. The system calculates power point tracking error and PTO power channel power point tracking error, and determines the dynamic priority weights of the driving power channel and PTO power channel based on PTO speed deviation, vehicle speed deviation, and battery SOC state. It optimizes fuel consumption, battery state of charge deviation, engine dynamic changes, driving power channel power point tracking error, and PTO power channel power point tracking error, while constraining the output capabilities of the engine, motor, battery, and PTO, to solve for the power distribution control quantity in the prediction time domain. Based on the power distribution control quantity, control commands are output to the engine controller, drive motor controller, PTO actuator, and battery management system. In specific implementation, see the appendix to the specification. Figure 1As shown, the state perception module collects throttle opening, vehicle speed, engine speed, PTO speed, operating resistance, SOC, and other state parameters; the load prediction module outputs future load prediction values ​​based on the above information; the dual-channel power decoupling module decomposes the total power demand into the power demand of the walking power channel and the power demand of the PTO power channel; the energy management optimization module performs optimization solutions according to the objective function and constraints; and the execution control module outputs control commands to the engine, drive motor, PTO system, and battery pack.

[0023] Status data acquisition and load forecasting: In each control cycle k, the state-aware module constructs a state vector: (1) in, For vehicle speed, For acceleration, Engine speed, For PTO rotation speed, This refers to the state of charge of the power battery. For traction resistance, For PTO torque, This refers to the tillage depth or working depth. The load forecasting model takes the load and state sequence within a historical window as input: (2) Output the predicted walk load and PTO load in the time domain: (3) The load prediction model can employ LSTM, GRU, TCN, or Transformer networks. During model training, offline data on working conditions such as rotary tillage, plowing, sowing, transportation, and baling are collected. Historical window sequences are used as inputs, and future load sequences are used as labels. Supervised training is performed using mean squared error loss. The loss function includes at least the walking load prediction error and the PTO load prediction error. After training, the model is deployed to the vehicle controller or an independent computing unit.

[0024] Dual-channel power decoupling: As per the instruction manual Figure 2 As shown, the dual-channel power decoupling module decomposes the total power demand into the power of the walking power channel and the power of the PTO power channel: (4) The power of the travel drive is calculated based on the predicted traction resistance, vehicle speed, and transmission efficiency: (5) in, This includes rolling resistance, gradient resistance, acceleration resistance, air resistance, and traction resistance term formed by predicted operating resistance. PTO power channel power is calculated based on predicted PTO torque and PTO speed. (6) By respectively and Setting power tracking error, limiting, and priority constraints can prioritize maintaining PTO operational stability during PTO load changes while limiting the rate of change of power in the travel power channel. Alternatively, when the travel load changes and the PTO speed is within the allowable range, the priority of the travel power channel can be increased, and the drive motor and power battery can compensate for the travel power gap, thereby limiting the impact of travel load changes on the PTO speed.

[0025] Energy management optimization and control output: The energy management optimization module constructs an optimization objective function within the prediction time domain H: (7) in For fuel consumption rate, For the target SOC, This represents the change in engine speed. This represents the change in engine power. , , , , These are the weighting coefficients. And... , It can adjust according to dynamic priority weights. Optimization constraints include: (8) (9) (10) (11) The optimization constraints include engine speed and torque constraints, motor speed and torque constraints, power battery SOC upper and lower limits constraints, power battery charging and discharging power constraints, PTO speed allowable deviation constraints, driving power channel power change rate constraints, PTO power channel power change rate constraints, and power balance constraints, where power battery power is positive for discharging and negative for charging. The resulting control quantities include target engine speed and torque, target drive motor torque, PTO motor target torque or PTO clutch control quantity, and power battery charging and discharging power limits. The execution control module sends these control quantities to the corresponding actuators via CAN bus or vehicle Ethernet.

[0026] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An energy management method for dual-channel power decoupling of tractors based on load prediction, characterized in that, Includes the following steps: Collect the tractor's operating status parameters, work load parameters, and power system status parameters to construct the current state vector; Based on the load sequence within a preset historical window and the state vector, the load prediction model outputs the walking load prediction sequence and PTO load prediction sequence in the prediction time domain. The power demand of the walking power channel is calculated based on the walking load prediction sequence, and the power demand of the PTO power channel is calculated based on the PTO load prediction sequence, so that the walking power channel and the PTO power channel each form independent power demand. The power requirements of the walking power channel are as follows calculate, It is determined at least by rolling resistance, gradient resistance, acceleration resistance, air resistance, and predicted traction operation resistance. For driving transmission efficiency; PTO power channel power requirements are based on calculate, To predict PTO torque, The target speed or measured speed for PTO. For PTO transmission efficiency; In the prediction time domain, the power tracking error of the driving power channel and the power tracking error of the PTO power channel are established respectively, and the dynamic priority weights of the driving power channel and the PTO power channel are determined according to the PTO speed deviation, vehicle speed deviation and power battery SOC state. When the PTO speed deviation is greater than the preset PTO speed deviation threshold, the priority of the PTO power channel is increased and the power change rate of the walking power channel is limited; when the walking speed deviation is greater than the preset vehicle speed deviation threshold and the PTO speed is within the allowable range, the priority of the walking power channel is increased and the driving motor and power battery compensate for the walking power gap. With fuel consumption, power battery state of charge deviation, engine dynamic changes, power tracking error of travel power channel and power tracking error of PTO power channel as optimization objectives, and with the output capabilities of engine, motor, power battery and PTO as constraints, the power distribution control quantity in the prediction time domain is solved. Control commands are output to the engine controller, drive motor controller, PTO actuator, and battery management system based on the power distribution control quantity.

2. The energy management method for dual-channel power decoupling of tractors based on load prediction according to claim 1, characterized in that, The operating status parameters include vehicle speed, acceleration, throttle opening, gear, engine speed, and PTO speed; the working load parameters include at least one of traction resistance, PTO torque, tillage depth, implement lifting status, soil moisture content, or working position; the power system status parameters include power battery SOC, battery allowable charge and discharge power, motor speed, motor torque, and current engine operating point.

3. The energy management method for dual-channel power decoupling of tractors based on load prediction according to claim 1, characterized in that, The state vector is constructed as follows: ； in For vehicle speed; Acceleration; Engine speed; PTO rotation speed; For traction resistance; PTO torque; This refers to the depth of tillage or working.

4. The energy management method for dual-channel power decoupling of tractors based on load prediction according to claim 2, characterized in that, The load prediction model employs a Long Short-Term Memory (LSTM) network, gated recurrent units (GNUs), temporal convolutional networks (TRNs), Transformer networks, or a combination of these models. It undergoes supervised training using historical window state sequences and future load sequences collected under conditions such as rotary tillage, plowing, sowing, and transportation. The loss function includes at least the walking load prediction error and the PTO (Pulsed Tolerance) load prediction error. The load prediction model uses... Input, Output Where N is the historical window length and H is the prediction time domain length. This is used to predict the traction resistance and PTO torque sequences over the time domain.

5. The energy management method for dual-channel power decoupling of a tractor based on load prediction according to claim 4, characterized in that, The dual-channel power decoupling satisfies the requirement that the power demand of the walking power channel and the power demand of the PTO power channel are calculated and constrained separately. Independent power tracking error, power limit, power change rate limit and priority constraint are set for the walking power channel and the PTO power channel respectively, so as to reduce the impact of PTO load change on walking speed and the impact of walking load change on PTO speed.

6. The method according to claim 5, characterized in that, The power tracking errors of the walking power channel and the PTO power channel are respectively: , ; in: Power required for the walking power channel; This represents the actual power of the walking power channel; Power required for PTO power channel; This represents the actual power of the PTO power channel, and is adjusted according to the aforementioned dynamic priority weights. and The penalty coefficient in the optimization objective function. Dynamic priority weights satisfy: when When the deviation exceeds the preset PTO speed deviation threshold, Increase; when Greater than the preset vehicle speed deviation threshold and When within the allowable range, make Increase; when the SOC is lower than the preset SOC lower limit protection value, limit the power battery discharge compensation power.

7. The energy management method for dual-channel power decoupling of tractors based on load prediction according to claim 1, characterized in that, The optimization objective function, with fuel consumption, battery state-of-charge deviation, engine dynamic changes, power tracking error of the driving power channel, and power tracking error of the PTO power channel as optimization objectives, is: ； in For fuel consumption rate, For the target SOC, This represents the change in engine speed. This represents the change in engine power. , , , , These are the weighting coefficients.

8. The energy management method for dual-channel power decoupling of tractors based on load prediction according to claim 1, characterized in that, The constraints include engine speed and torque constraints, motor speed and torque constraints, upper and lower limits of power battery SOC constraints, power battery charging and discharging power constraints, allowable deviation constraints of PTO speed, power change rate constraints of the driving power channel, power change rate constraints of the PTO power channel, and power balance constraints, wherein the power battery power is positive when discharging and negative when charging.

9. A tractor dual-channel power decoupling energy management system based on load prediction, characterized in that, It includes: The module includes a state awareness module, a load prediction module, a dual-channel power decoupling module, an energy management optimization module, and an execution control module. The state perception module is used to collect tractor operating state parameters, work load parameters, and power system state parameters; the load prediction module is used to output the travel load prediction sequence and PTO load prediction sequence in the prediction time domain; the dual-channel power decoupling module is used to calculate the power demand of the travel power channel and the power demand of the PTO power channel respectively; the energy management optimization module is used to solve the power distribution control quantities of the engine, motor, PTO, and power battery under constraints; and the execution control module is used to send the power distribution control quantities to the engine controller, drive motor controller, PTO actuator, and battery management system.

10. The energy management system for tractor dual-channel power decoupling based on load prediction according to claim 9, characterized in that, The PTO actuator includes at least one of a PTO motor, a PTO clutch, a PTO brake, or a PTO power converter, and the execution control module can perform closed-loop control of the PTO output torque or PTO output speed according to the PTO channel power requirements.

Images