A double planetary hybrid tractor gearbox configuration design method based on topological feature clustering
By constructing a topology feature clustering method for dual planetary hybrid tractor gearboxes, a topology with low-power cycles and controllable motor overspeed risk was selected, achieving multi-condition adaptability and system stability, and solving the problems existing in the current design method.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NANJING INST OF RAILWAY TECH
- Filing Date
- 2026-05-29
- Publication Date
- 2026-07-03
AI Technical Summary
Existing hybrid tractor gearbox design methods fail to effectively balance multi-condition adaptability, low-power cycles, and motor overspeed risks, and lack a unified structural-level evaluation standard, resulting in high design complexity and increased verification costs.
A design method based on topological feature clustering is adopted to construct the static topological space of the dual planetary hybrid tractor gearbox. By screening topological feature indicators and clustering analysis, combined with clutch and brake optimization, multi-condition adaptability and low-power cycle characteristics are achieved.
It significantly improves topology screening efficiency, reduces design complexity and verification costs, ensures system stability and efficiency, and adapts to complex and ever-changing agricultural operating conditions.
Smart Images
Figure CN122333646A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of hybrid tractor transmission system design technology, specifically relating to a dual planetary hybrid tractor gearbox configuration design method based on topological feature clustering. Background Technology
[0002] With the continuous improvement of agricultural mechanization and increasingly stringent environmental protection requirements, hybrid power transmission systems have received widespread attention for their application in high-power tractors due to their excellent fuel economy and power performance. The power split architecture based on a planetary gear mechanism possesses continuous speed regulation, multi-power source coupling, and engine operating point optimization capabilities, making it particularly suitable for various agricultural operations such as transportation, plowing, and rotary tillage.
[0003] However, agricultural tractors are characterized by complex operating conditions, large load fluctuations, and the need to simultaneously meet the dual power requirements of traction drive and PTO (Power Towing) operations. Traditional design methods based on fixed transmission ratios or single optimization objectives struggle to simultaneously balance power performance, economy, and system stability. Existing hybrid power topology generation methods can construct a large-scale design space through automated enumeration searches, but the sheer number of candidate configurations is enormous, and most solutions fail to meet practical engineering needs, leading to a significant increase in subsequent analysis and verification costs.
[0004] Specifically, to reduce design complexity, researchers have proposed rapid dimensionality reduction methods such as kinematic feasibility testing, pattern analysis, and isomorphic screening, as well as further screening methods based on indicators such as fuel economy and power performance. However, these methods largely rely on empirical rules and lack unified and interpretable structural-level evaluation standards. Performing dynamic simulations or experimental verification on a large number of candidate configurations is costly and fails to meet the needs of rapid iteration and multi-condition adaptability in agricultural machinery design. Furthermore, existing design methods fail to fully consider the tractor-specific PTO output requirements and the efficiency degradation and motor overspeed risks caused by internal power cycles.
[0005] Therefore, there is an urgent need to develop a dual planetary hybrid tractor gearbox configuration design method that can quickly identify potential topology schemes in the early stages of design, while taking into account multi-condition adaptability, low power cycle and low risk of motor overspeed. Summary of the Invention
[0006] The purpose of this invention is to provide a dual planetary hybrid tractor gearbox configuration design method based on topological feature clustering, which aims to solve the problems of low efficiency in topology design, lack of unified structural-level evaluation standards for screening, and difficulty in simultaneously meeting the requirements of multiple working conditions and system stability in the existing technology.
[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for designing the configuration of a dual planetary hybrid tractor gearbox based on topological feature clustering, comprising the following steps: S1. Constructing the complete static topology space of the dual planetary gearbox hybrid transmission: Under the port uniqueness constraint and component single port constraint, the system enumerates the port connection relationship between the engine E, the two motors MG1 / MG2 and the dual planetary gearbox PG1 / PG2, and combines the three typical positions of the PTO power take-off point (engine shaft E, intermediate shaft Lm, output shaft Lo) to generate a full combinatorial search space containing 108 candidate topologies. S2. Establish a quantitative model of topological features: Based on the analytical kinematic relationship of the planetary arrangement, construct a global topological constraint matrix, introduce unified and standardized working condition constraints and the principle of solving the minimum motor speed norm, define and calculate four structural topological feature indices, namely PTO-vehicle speed coupling strength, engine decoupling difficulty, traction motor torque increase potential and motor speed pressure. S3. Clustering analysis and preliminary screening based on topological features: The range of four topological feature indicators is normalized to construct a topological feature vector. The K-means clustering algorithm initialized with K-means++ is used to classify 108 candidate topologies. The optimal number of clusters is determined by combining the elbow method with the silhouette coefficient. Then, the candidate topology with the best comprehensive performance is selected from each class based on the multi-index comprehensive evaluation function. S4. Power Cycling and Overspeed Risk Special Assessment: Calculate the speed hedging index and hedging strength index of each candidate topology, and select the basic topology with low power cycling trend and controllable motor overspeed risk by combining the motor speed amplitude. S5. Secondary topology optimization design: Based on the basic topology, switching elements such as clutch brake and clutch are introduced. By adjusting the power source connection relationship and adding a mode switching mechanism, the risk of motor overspeed is reduced while maintaining low power cycle characteristics, and multiple working modes can be switched.
[0008] As a design method for a dual planetary hybrid tractor gearbox configuration based on topological feature clustering according to the present invention, preferably, the port uniqueness constraint in S1 is that each port of the same planetary gear set is only allowed to connect to one mechanical shaft at any given time; the component single port constraint is that the engine and each motor are only allowed to connect to one mechanical port; the enumeration of the dual planetary gear set port connection relationship is specifically as follows: the port mapping set of the front-stage planetary gear set PG1 is {E, MG2, Lm}, the port mapping set of the rear-stage planetary gear set PG2 is {Lm, MG1, Lo}, and the two planetary gear sets are cascaded through the intermediate shaft Lm.
[0009] As a design method for a dual-planetary hybrid tractor gearbox configuration based on topological feature clustering according to the present invention, preferably, the four topological feature indices in S2 are defined as follows: PTO-vehicle speed coupling strength: The equivalent change in PTO speed when the output shaft speed changes unit under standard operating conditions, taking the absolute value to eliminate the influence of rotation direction; Engine decoupling difficulty: Under the normalization constraint of output shaft speed, the equivalent speed norm of the motor that satisfies the kinematic constraints of the planetary gear set; Traction motor torque enhancement potential: the speed amplitude of traction motor MG2 under normalized output speed conditions; Motor speed pressure: The larger speed amplitude of the two motors when the tractor outputs its maximum speed.
[0010] As a design method for a dual planetary hybrid tractor gearbox configuration based on topological feature clustering according to the present invention, preferably, the method for determining the optimal number of clusters in S3 is as follows: record the intra-cluster squared error SSE under different numbers of clusters, determine the main inflection point by the elbow method, calculate the average profile coefficient in the elbow neighborhood, select the number of clusters with the largest profile coefficient as the optimal number of clusters, and finally determine the optimal number of clusters as 4.
[0011] As a design method for a dual-planetary hybrid tractor gearbox configuration based on topological feature clustering according to the present invention, preferably, the multi-index comprehensive evaluation function in S3 is: S = w1f1 + w2f2 + w3f3 + w4f4; Where f1 is the normalized PTO-vehicle speed coupling strength, f2 is the normalized engine decoupling difficulty, f3 is the normalized traction motor torque increase potential, and f4 is the normalized motor speed pressure; the weighting coefficients w1=0.35, w2=0.30, w3=0.20, w4=0.15, and w1+w2+w3+w4=1; the smaller the comprehensive score S, the better the topology comprehensive performance.
[0012] As a design method for a dual planetary hybrid tractor gearbox configuration based on topological feature clustering in this invention, preferably, the speed offset index in S4 is defined as the product sign of the speeds of the two motors. When the product is negative, it indicates that there is speed offset, i.e., a power cycle trend. The offset intensity index is the cumulative average value of the speed offset ratio within the entire operating range. The larger the index, the greater the potential intensity of the internal cycle power.
[0013] As a design method for a dual planetary hybrid tractor gearbox configuration based on topological feature clustering according to the present invention, preferably, the basic topology in S5 is a G22 configuration, characterized in that: the PTO is directly connected to the engine shaft, and the rear planetary gear set PG2 adopts a power splitting-combining structure with the planet carrier as the output port, the sun gear connected to MG1, and the gear ring connected to the intermediate shaft Lm.
[0014] As a design method for a dual-planetary hybrid tractor gearbox configuration based on topological feature clustering according to the present invention, preferably, the secondary topology optimization in S5 specifically includes: A clutch brake CB is installed between the sun gears of MG1 and PG2; Add a clutch C1 between the engine and the MG1 motor; Add a clutch C2 between the MG1 motor and the PTO; Add brake B2 between the MG2 motor and the PG1 sun gear; The fixed PG1 gear ring makes the front planetary gear set form a fixed-axis gear system, achieving speed reduction and torque increase.
[0015] As a design method for a dual planetary hybrid tractor gearbox configuration based on topological feature clustering, the optimized gearbox configuration can preferably achieve four working modes: pure electric drive, series drive, hybrid drive, and engine direct drive. The mode switching is achieved by controlling the engagement and disengagement states of each clutch and brake.
[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention constructs a complete static topology space for dual planetary sorting systems containing 108 candidate topologies, comprehensively covering all possible connection methods under given component sets and port constraints, avoiding the omission of excellent topology schemes, and providing a sufficient selection basis for configuration design.
[0017] This invention proposes four structural-level topology characteristic indicators, which enable quantitative evaluation of the PTO coupling capability, engine decoupling degree, traction torque enhancement potential, and motor speed pressure of different topologies at the structural level in the early design stage, without the need for costly dynamic simulation, thus significantly improving the efficiency of topology selection.
[0018] This invention classifies topologies using K-means clustering analysis, which can identify potential categories of different topologies in terms of structure and function. Combined with a multi-index comprehensive evaluation function, representative topologies are selected from each category, which not only ensures the comprehensiveness of the selection but also greatly reduces the workload of subsequent analysis.
[0019] This invention, through a specialized assessment of power cycling and overspeed risk, can accurately identify basic topologies with low-power cycling trends and controllable motor overspeed risk, effectively solving the efficiency reduction and motor overspeed problems caused by internal power cycling in existing design methods. This invention optimizes the basic topology by introducing switching elements to adjust the power source connection relationship and adding a mode switching mechanism. While maintaining low-power cycling characteristics, it reduces the risk of motor overspeed and realizes multiple working mode switching, which can better adapt to the complex and ever-changing working conditions of tractors.
[0020] This invention verifies the optimized gearbox configuration through system-level simulation under three typical operating conditions: transportation, plowing, and rotary tillage, ensuring the engineering applicability and reliability of the design scheme. Attached Figure Description
[0021] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram illustrating the steps of a dual-planetary hybrid tractor gearbox configuration design method based on topological feature clustering.
[0022] Figure 2 This is a schematic diagram showing the results of calculating the speed hedging index, hedging strength index, and motor speed amplitude of eight candidate topologies in step 4 of the dual planetary hybrid tractor gearbox configuration design method based on topological feature clustering. Detailed Implementation
[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. 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.
[0024] Please see Figure 1-2 The present invention provides the following technical solution: a method for designing the configuration of a dual planetary hybrid tractor gearbox based on topological feature clustering, the specific implementation steps of which are as follows:
[0025] Step 1: Construct the complete static topology space of the dual planetary gearbox hybrid transmission Under the port uniqueness constraint (each port of the same planetary gear set is only allowed to connect to one mechanical shaft at any given time) and the component single port constraint (the engine and each motor are only allowed to connect to one mechanical port), the system enumerates the port connection relationships of the engine E, the two motors MG1 / MG2 and the double planetary gear set PG1 / PG2.
[0026] The front-stage planetary gear set PG1 is abstracted as a three-port power coupling node with a sun gear S1, a ring gear R1, and a planet carrier C1. The set of components to be connected is {E, MG2, Lm}, and the total number of port mappings is 3! = 6. The rear-stage planetary gear set PG2 is similarly abstracted as a three-port power coupling node with a sun gear S2, a ring gear R2, and a planet carrier C2. The set of components to be connected is {Lm, MG1, Lo}, and the total number of port mappings is also 3! = 6. The two planetary gear sets are cascaded through an intermediate shaft Lm, therefore the total number of core mechanical topologies is 6 × 6 = 36.
[0027] Combining the three typical locations of the PTO power take-off point (engine shaft E, intermediate shaft Lm, and output shaft Lo), a full combinatorial search space containing \(36×3=108\) candidate topologies is finally generated.
[0028] Step 2: Establish a topological feature quantification model Based on the analytical kinematics of classical planetary gear sets, for the k-th simple planetary gear set, the angular velocities of its sun gear S, ring gear R, and planet carrier C satisfy the following: W S +k k W R -(1+k k W C =0; Where, k k =z R / z S The structural transmission parameters of the k-th planetary gear set are uniquely determined by the gear ratio.
[0029] Introducing the global mechanical axis velocity vector W=[W E W MG1 W MG2 W Lm W Lo ] T Substituting the port connection relationships into the above kinematic equations, we obtain a set of linear constraint equations for the global velocity variables: A(x) W =0; Where A(x) is a constraint matrix uniquely determined by the topological structure x.
[0030] To ensure comparability between different topologies, a unified normalized operating condition constraint (setting the vehicle output shaft speed to a unit value) and a minimum motor speed norm solution principle are introduced. That is, under the conditions of satisfying the planetary gear set kinematic constraints and normalized operating conditions, the feasible solution with the minimum motor speed amplitude is selected. min‖ω MG1 ,ω MG2 ‖2; st A(x) ω =0, ωLo =1; Based on the above normalized model, four structural-level topological feature indices are defined and calculated: 1. PTO-vehicle speed coupling strength: f1=|dw PTO / dw Lo | reflects the equivalent change in PTO speed when the output shaft speed changes by a unit; 2. Difficulty of engine decoupling: f2 = ||w MG1 ,w MG2 ‖2, Motor equivalent speed norm under output shaft speed normalization constraint; 3. Traction motor torque increase potential: f3=|w MG2 |, the speed amplitude of traction motor MG2 under normalized output speed conditions; 4. Motor speed and pressure: f4 = max(|w MG1 ,max|,|w MG2 ,max|), is the larger speed amplitude of the two motors when the tractor outputs its maximum speed.
[0031] Step 3: Cluster analysis and preliminary screening based on topological features The optimal number of clusters was determined by combining the elbow method with the silhouette coefficient: the intra-cluster squared error (SSE) was recorded under different numbers of clusters. When the number of clusters increased to 4, the downward trend of SSE slowed down significantly, showing a typical "elbow" characteristic. At the same time, the average silhouette coefficient was the largest in the elbow neighborhood when the number of clusters was 4. Therefore, the optimal number of clusters was determined to be 4.
[0032] Step 4: Special Assessment of Power Cycle and Overspeed Risk The speed offset index is defined as the product of the speeds of the two motors: I 对冲 =W MG1 *W MG1 ; When I 对冲 When the value is less than 0, it indicates that the two motors rotate in opposite directions, resulting in speed conflict, i.e., power cycling.
[0033] Step 5: Secondary topology optimization design The basic topology G22 is characterized by: the PTO being directly connected to the engine shaft; the rear planetary gear set PG2 using a power splitting-combining structure with the planet carrier as the output port, the sun gear connected to MG1, and the ring gear connected to the intermediate shaft Lm. This structure allows MG1 to be used only to balance the power difference between the mechanical branches and the output, thus avoiding systemic power conflict with MG2.
[0034] To address the issue of excessively high MG1 rotational speed in the G22 configuration, the following secondary topology optimization is performed: 1. A clutch brake CB is installed between the sun gears of MG1 and PG2 to achieve power decoupling when MG1 is not engaged in operation; 2. Add a clutch C1 between the engine and the MG1 motor so that the engine power can be directly transmitted to the MG1 shaft; 3. Add clutch C2 between MG1 motor and PTO to achieve independent control of PTO; 4. Add brake B2 between MG2 motor and PG1 sun gear to fix PG1 gear ring so that the front planetary gear set forms a fixed-axis gear system to achieve speed reduction and torque increase; 5. Add a main reducer clutch C3 and a wheel-side reducer clutch C4 to achieve switching between different transmission ratios.
[0035] The optimized gearbox configuration can achieve the following four operating modes: 1. Pure electric drive mode: Clutch C1 is disengaged, clutch brake CB is engaged, brake B2 is disengaged, and MG2 drives the vehicle independently. 2. Series drive mode: Clutch C1 is engaged, clutch brake CB is disengaged, engine drives MG1 to generate electricity, and MG2 drives the vehicle; 3. Hybrid drive mode: Clutch C1 is engaged, clutch brake CB is engaged, and the engine and two electric motors drive the vehicle together. 4. Engine direct drive mode: Clutch C1 is engaged, clutch brake CB is engaged, brake B2 is engaged, and the engine directly drives the vehicle.
[0036] Step 6: System-level simulation verification A MATLAB / Simscape system-level simulation model was established, and the following parameters were set according to the power matching requirements of a 220-horsepower hybrid tractor: Engine rated power: 90kW, economical operating speed: 2300rpm; MG1 rated power: 55kW, maximum permissible speed: 4000rpm; MG2 rated power: 75kW, maximum permissible speed: 2400rpm; Characteristic parameters of the front and rear planetary array structure: k1=2.9, k2=0.7; Maximum output shaft speed: 2300 rpm.
[0037] Multi-condition verification was conducted under three typical operating conditions: transportation, plowing, and rotary tillage. Simulation results show that: 1. The optimized transmission configuration can achieve accurate vehicle speed tracking under all operating conditions, with a tracking error of less than 2%; 2. Both motors are operating within a safe speed range, with no risk of overspeeding; 3. The internal power circulation of the system is less than 5%, and the transmission efficiency is significantly improved; 4. Compared with traditional mechanical transmission structures, fuel consumption is reduced by 15%-20%.
[0038] The control method of the present invention follows a flow principle and specifically includes the following stages: Phase 1: Topology Space Construction Phase The core of this stage is to generate a complete static topology space for the dual planetary gearbox hybrid transmission. First, the core components of the system (engine E, two electric motors MG1 / MG2, dual planetary gearboxes PG1 / PG2, and PTO) and basic constraints (port uniqueness constraints and component single-port constraints) are identified. Then, the port mapping relationships between the preceding planetary gearbox PG1 and the following planetary gearbox PG2 are enumerated, and the two planetary gearboxes are cascaded through the intermediate shaft Lm, resulting in 36 core mechanical topologies. Finally, combining the three typical power take-off positions of the PTO, a full combinatorial search space containing 108 candidate topologies is generated.
[0039] The principle of this stage is to enumerate all possible port connection relationships through the system to ensure that no potential excellent topology schemes are overlooked, thus providing a sufficient foundation for subsequent screening and optimization.
[0040] Phase Two: Topological Feature Modeling Phase The core of this stage is to establish a structural-level topology feature quantification model to enable quantitative comparison of different topologies. First, based on the analytical kinematics of the planetary gear set, a global topology constraint matrix is constructed, transforming the structural differences between topologies into differences in the constraint matrix. Then, a unified and normalized operating condition constraint and a minimum motor speed norm solution principle are introduced to eliminate system-scale degrees of freedom and establish a unified reference state. Finally, four structural-level topology feature indices are defined and calculated to quantitatively evaluate topology performance from four dimensions: PTO coupling capability, engine decoupling degree, traction torque enhancement potential, and motor speed pressure.
[0041] The principle behind this stage is that differences in topology essentially stem from differences in the connection relationships between the power source and the planetary gear set ports, leading to variations in the system's kinematic characteristics. By extracting structural-level features that reflect these kinematic characteristics, topology performance can be rapidly evaluated early in the design process without the need for costly dynamic simulations.
[0042] Phase 3: Cluster Analysis and Preliminary Screening The core of this stage is to achieve topology classification and preliminary screening through cluster analysis. First, the range of four topological characteristic indicators is normalized to eliminate the interference of dimensional effects on the clustering results. Then, a K-means clustering algorithm initialized with K-means++ is used to classify 108 candidate topologies. The optimal number of clusters is determined to be 4 using the elbow method combined with the silhouette coefficient. Finally, based on a multi-index comprehensive evaluation function, the top two configurations with the best overall performance in each cluster are selected as representative candidates, resulting in a total of 8 candidate topologies.
[0043] The principle behind this stage is that topologies with similar structural features also have functional similarities. Cluster analysis can divide a large number of candidate topologies into a few categories, each representing a typical structural-functional pattern. Selecting representative topologies from each category ensures comprehensive screening while significantly reducing the workload of subsequent analysis.
[0044] Phase Four: Specialized Assessment and Secondary Optimization The core of this phase is to select a basic topology with low power cycling trends and controllable motor overspeed risk, and then perform secondary optimization. First, the speed offset index and offset strength index of each candidate topology are calculated to assess its power cycling risk; simultaneously, the overspeed risk is assessed in conjunction with the motor speed amplitude. Considering both power cycling characteristics and motor speed safety, G22 is selected as the basic topology. Then, based on the basic topology, switching elements such as clutches and brakes are introduced, the power source connection relationships are adjusted, and a mode switching mechanism is added. This reduces the motor overspeed risk while maintaining low power cycling characteristics and enables multi-mode switching.
[0045] The principle behind this phase is that power cycling and motor overspeed are key issues affecting the efficiency and reliability of hybrid powertrains. These issues can be accurately identified through specialized assessments, and then addressed specifically through secondary topology optimization, while retaining the advantages of the basic topology.
[0046] Phase 5: System-level simulation verification phase The core of this phase is to verify the engineering applicability and reliability of the optimized gearbox configuration. A MATLAB / Simscape system-level simulation model is established to conduct multi-condition verification under three typical operating conditions: transportation, plowing, and rotary tillage. The focus is on evaluating vehicle speed tracking performance, motor speed safety, power cycle characteristics, and fuel economy. Based on the simulation results, the design scheme is adjusted and improved to ensure it meets actual engineering requirements.
[0047] The principle of this stage is that system-level simulation can simulate the operation of the gearbox under various working conditions in a virtual environment, comprehensively evaluate its performance, promptly identify and resolve problems in the design, and reduce actual development costs and risks.
[0048] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for designing a two planetary hybrid power tractor gearbox configuration based on topology feature clustering, characterized in that, Includes the following steps: S1. Constructing the complete static topology space of the dual planetary gearbox hybrid transmission: Under the port uniqueness constraint and component single port constraint, the system enumerates the port connection relationship between the engine E, the two motors MG1 / MG2 and the dual planetary gearbox PG1 / PG2, and combines the three typical positions of the PTO power take-off point (engine shaft E, intermediate shaft Lm, output shaft Lo) to generate a full combinatorial search space containing 108 candidate topologies. S2. Establish a quantitative model of topological features: Based on the analytical kinematic relationship of the planetary arrangement, construct a global topological constraint matrix, introduce unified and standardized working condition constraints and the principle of solving the minimum motor speed norm, define and calculate four structural topological feature indices, namely PTO-vehicle speed coupling strength, engine decoupling difficulty, traction motor torque increase potential and motor speed pressure. S3. Clustering analysis and preliminary screening based on topological features: The range of four topological feature indicators is normalized to construct a topological feature vector. The K-means clustering algorithm initialized with K-means++ is used to classify 108 candidate topologies. The optimal number of clusters is determined by combining the elbow method with the silhouette coefficient. Then, the candidate topology with the best comprehensive performance is selected from each class based on the multi-index comprehensive evaluation function. S4. Power Cycling and Overspeed Risk Special Assessment: Calculate the speed hedging index and hedging strength index of each candidate topology, and select the basic topology with low power cycling trend and controllable motor overspeed risk by combining the motor speed amplitude. S5. Secondary topology optimization design: Based on the basic topology, switching elements such as clutch brake and clutch are introduced. By adjusting the power source connection relationship and adding a mode switching mechanism, the risk of motor overspeed is reduced while maintaining low power cycle characteristics, and multiple working modes can be switched.
2. The method for configuration design of a two-row planetary hybrid power take-off tractor gearbox based on topology feature clustering according to claim 1, characterized in that: The port uniqueness constraint in S1 means that each port of the same planetary gear set is only allowed to connect to one mechanical shaft at any given time; the component single-port constraint means that the engine and each motor are only allowed to connect to one mechanical port; the enumeration of the dual planetary gear set port connection relationship is as follows: the port mapping set of the front-stage planetary gear set PG1 is {E, MG2, Lm}, the port mapping set of the rear-stage planetary gear set PG2 is {Lm, MG1, Lo}, and the two planetary gear sets are cascaded through the intermediate shaft Lm.
3. The method for configuration design of a two-row planetary hybrid power take-off tractor gearbox based on topology feature clustering according to claim 1, characterized in that: The four topological feature indicators in S2 are defined as follows: PTO-vehicle speed coupling strength: The equivalent change in PTO speed when the output shaft speed changes unit under standard operating conditions, taking the absolute value to eliminate the influence of rotation direction; Engine decoupling difficulty: Under the normalization constraint of output shaft speed, the equivalent speed norm of the motor that satisfies the kinematic constraints of the planetary gear set; Traction motor torque enhancement potential: the speed amplitude of traction motor MG2 under normalized output speed conditions; Motor speed pressure: The larger speed amplitude of the two motors when the tractor outputs its maximum speed.
4. The method for configuration design of double planetary hybrid power tractor gearbox based on topology feature clustering according to claim 1, characterized in that: The method for determining the optimal number of clusters in S3 is as follows: record the intra-cluster squared error SSE under different numbers of clusters, determine the main inflection points by the elbow method, calculate the average profile coefficient in the elbow neighborhood, select the cluster number with the largest profile coefficient as the optimal number of clusters, and finally determine the optimal number of clusters as 4.
5. The method for configuration design of double planetary hybrid power tractor gearbox based on topology feature clustering according to claim 1, characterized in that: The multi-index comprehensive evaluation function in S3 is: S = w1f1 + w2f2 + w3f3 + w4f4; Where f1 is the normalized PTO-vehicle speed coupling strength, f2 is the normalized engine decoupling difficulty, f3 is the normalized traction motor torque increase potential, and f4 is the normalized motor speed pressure; the weighting coefficients w1=0.35, w2=0.30, w3=0.20, w4=0.15, and w1+w2+w3+w4=1; the smaller the comprehensive score S, the better the topology comprehensive performance.
6. The method for configuration design of double planetary hybrid power tractor gearbox based on topology feature clustering according to claim 1, characterized in that: The speed offset index in S4 is defined as the product of the speeds of the two motors. When the product is negative, it indicates that there is speed offset, i.e., a power cycle trend. The offset intensity index is the cumulative average of the speed offset ratio within the entire operating range. The larger the index, the greater the potential intensity of the internal cycle power.
7. The method for configuration design of a two-row planetary hybrid power take-off tractor gearbox based on topology feature clustering according to claim 1, characterized in that: The basic topology in S5 is a G22 configuration, characterized by: PTO being directly connected to the engine shaft, and the rear planetary gear set PG2 using a power splitting-combining structure with a planetary carrier as the output port, the sun gear connected to MG1, and the gear ring connected to the intermediate shaft Lm.
8. The method for designing a two planetary hybrid power tractor gearbox configuration based on topology feature clustering as claimed in claim 1, wherein: The secondary topology optimization in S5 specifically includes: A clutch brake CB is installed between the sun gears of MG1 and PG2; Add a clutch C1 between the engine and the MG1 motor; Add a clutch C2 between the MG1 motor and the PTO; Add brake B2 between the MG2 motor and the PG1 sun gear; The fixed PG1 gear ring makes the front planetary gear set form a fixed-axis gear system, achieving speed reduction and torque increase.
9. The method for designing a two planetary hybrid power take-off transmission configuration based on topological feature clustering according to claim 8, characterized in that: The optimized gearbox configuration can achieve four working modes: pure electric drive, series drive, hybrid drive, and engine direct drive. The mode switching is achieved by controlling the engagement and disengagement of each clutch and brake.