A hybrid shift schedule optimization method based on working condition data

Abstract

The application discloses a hybrid gear shift rule optimization method based on working condition data and belongs to the technical field of automobile automatic variable speed driving systems; the method comprises the following steps: acquiring typical working condition information of a target vehicle model; generating an optimal control trajectory according to the typical working condition information and in combination with an established value function; determining a global optimal control law according to the optimal control trajectory; obtaining a gear shift point in a coordinate system according to the global optimal control law; clustering the gear shift point by using a K-means clustering algorithm to obtain a clustering center; fitting and generating a gear shift curve according to the clustering center; and determining an optimal gear shift timing based on the gear shift curve and performing a gear shift operation. The application provides a control method of a hybrid optimal gear shift rule with high engineering application value, strong practicability and simultaneous consideration of multiple working conditions for a vehicle driving system matched with a multi-gear transmission.

Description

Technical Field

[0001] This invention belongs to the technical field of automotive automatic transmission drive systems, specifically relating to a method for optimizing hybrid shifting patterns based on operating condition data. Background Technology

[0002] Traditional gasoline-powered vehicles typically require a transmission to adjust engine output characteristics and meet the vehicle's power demands. With the rise of new energy vehicles, especially pure electric vehicles, some passenger cars and light vehicles can be directly driven by electric motors. However, for medium and heavy-duty commercial vehicles such as highway buses and dump trucks operating under complex conditions, a transmission scheme combining an electric motor and a multi-speed transmission is still necessary. Under different operating conditions, different gears are switched to change the operating point of the drive motor, thereby meeting the high torque requirements of the vehicle during start-up and hill climbing. Therefore, the selection of the shift pattern has a significant impact on the output characteristics of the drive system, directly affecting key indicators such as vehicle power and fuel economy.

[0003] Based on the different parameters set for the shifting pattern, traditional shifting patterns can be divided into single-parameter shifting patterns, two-parameter shifting patterns, and three-parameter shifting patterns. Among them, the two-parameter shifting pattern, which uses vehicle speed and accelerator pedal opening as control parameters, is widely used in the engineering field.

[0004] The most basic method for formulating shifting rules is to realize vehicle gear switching based on logical threshold rules based on subjective experience. Obviously, this method is not adaptable to driving conditions.

[0005] Another approach involves using data such as the efficiency diagram and external characteristics of the motor or engine to establish two shifting rules targeting power and economy, respectively. In practical applications, the two rules are switched via an external button or built-in rules. Although this method improves adaptability to different operating conditions, the selection of the switching timing becomes a new problem. To achieve better results, complex online recognition of driver intentions or accurate and timely judgment by the driver is required.

[0006] Some methods collect vehicle information, predicted operating conditions, and driving expectations, and perform online optimization calculations. These typically require neural network algorithms to perform real-time calculations on large amounts of data to determine the optimal gear position. While this method can theoretically handle various complex operating conditions, it places extremely high demands on computing power and information storage, and the output results are uncertain, making it difficult to implement in currently mass-produced vehicles.

[0007] The invention patent CN108333921A discloses a "Method for Optimizing Vehicle Gear Shifting Patterns Based on Dynamic Programming Algorithm." This patent's technology involves establishing a MAP (Modular Map) of engine fuel consumption characteristics. Given three driving cycle conditions and vehicle parameter data, it establishes a dynamic equation with gear position as the state variable, gear shift control command as the control variable, and overall fuel consumption for the entire driving cycle as the optimal objective function. It compares the throttle opening data obtained from the dynamic programming-based gear shifting pattern and the power-based gear shifting pattern under the same conditions, and adjusts the dynamic programming-based gear shifting curve to make the throttle opening as close as possible to the power-based gear shifting pattern while ensuring fuel economy, thus guaranteeing vehicle power. The final result is a gear shifting pattern curve that satisfies both power and fuel economy. Although this method solves for the optimal fuel-efficient gear shifting pattern offline using engine and driving cycle data, the adjustment process combined with power requirements still requires repeated manual calibration based on experience, and recalculation is necessary whenever a new set of driving cycle data appears.

[0008] The invention patent CN116201890A discloses a "Design Method for Adaptive Gear Shifting Law of Multi-Gear AMT Pure Electric City Bus". This patent's technology, based on vehicle parameters and characteristics, optimizes the gear shifting speed under different accelerator pedal openings using a multi-objective particle swarm optimization algorithm to establish a MOPSO gear shifting law. Simultaneously, it constructs a dual-input / single-output fuzzy controller, using actual load and acceleration changes of the city bus as input and the gear shifting point speed adjustment as output, to adjust the gear shifting law. While this method can balance power and economy in gear shifting, it requires online solving of the multi-objective particle swarm and fuzzy control algorithms, making real-time performance difficult to guarantee on typical vehicles.

[0009] As can be seen, the methods described above, or combinations thereof, all present certain challenges in engineering applications. Especially when dealing with the complex environments and variable operating conditions of commercial vehicles, it is difficult to simultaneously achieve optimal performance in terms of drive capability, gear shifting control, and real-time algorithm performance. Summary of the Invention

[0010] The purpose of this invention is to provide an efficient method for optimizing hybrid shifting patterns based on operating condition data.

[0011] To address the aforementioned technical problems, this invention provides a method for optimizing hybrid shifting patterns based on operating condition data, comprising the following steps:

[0012] Obtain typical operating condition information for the target vehicle model;

[0013] Based on typical operating condition information and the established value function, the optimal control trajectory is generated.

[0014] Determine the globally optimal control law based on the optimal control trajectory;

[0015] Based on the global optimal control law, the shift point is obtained in the coordinate system;

[0016] The shift points are clustered using the K-means clustering algorithm to obtain cluster centers.

[0017] The shift curve is generated by fitting the cluster centers;

[0018] Based on the shift curve, the optimal shift point is determined and the shift operation is executed.

[0019] Preferably, based on typical operating condition information and the established value function, the optimal control trajectory is generated, specifically including the following steps:

[0020] Establish a value function based on the state variables and control variables of typical operating conditions;

[0021] Based on the value function, the optimal control trajectory is generated using a dynamic programming algorithm.

[0022] Preferably, the formula for the value function is:

[0023]

[0024] In the formula: N is the total number of steps in the discrete driving condition; L is the instantaneous cost function; Ω is the adjustable penalty factor for shift frequency; J is the value function; x k It is a state variable; u k It is a control variable; P bat It refers to battery power.

[0025] Preferably, the state variables include the battery state of charge (SOC) and the gearbox gear.

[0026] The control variables include the requested torque Trq and the transmission shift command shift;

[0027] The formulas for state variables and control variables are:

[0028]

[0029] Preferably, the constraint condition of the value function is:

[0030]

[0031] Preferably, the formula for calculating the optimal control trajectory is as follows:

[0032]

[0033] Preferably, the shift point is obtained in the coordinate system according to the global optimal control law, specifically including the following steps:

[0034] According to the global optimal control law, the shift point is obtained in the corresponding vehicle speed-pedal opening or vehicle speed-pedal opening-gradient coordinate system.

[0035] Preferably, the shift points are clustered using the K-means clustering algorithm to obtain cluster centers, specifically including the following steps:

[0036] Determine the number of cluster centers, k;

[0037] Randomly select k points as initial cluster centers, representing k clusters;

[0038] Calculate the Eulerian distance between each shift point and each initial cluster center, and assign each shift point to the cluster with the closest Eulerian distance;

[0039] Based on the shift point of each cluster, recalculate the new cluster centers;

[0040] This continues until the deviation between the previous cluster center and the current cluster center is less than the deviation threshold.

[0041] Preferably, the formula for calculating the weighting function is:

[0042]

[0043] In the formula: f v (x) is an sigmoid function. It is a combined Gaussian function, which represents a weighted function of the proportions of speed and pedal position at a certain working condition; λ, a k b k β is an adjustment parameter;

[0044] The formula for calculating the Euler distance is:

[0045]

[0046] In the formula: D(x, μ) is the cluster distance considering weights.

[0047] Preferably, the formula for recalculating the new cluster centers is:

[0048]

[0049] Where: μ′ j It is a new cluster center.

[0050] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0051] This invention provides a control method for automotive drive systems matched with multi-speed transmissions, which has high engineering application value, strong practicality, and can simultaneously take into account the hybrid optimal shifting law under multiple operating conditions.

[0052] This invention can solve for the optimal shift point under various typical vehicle operating conditions using an offline optimal algorithm. At the same time, it fits a weight factor based on the probability distribution of various operating conditions. When performing clustering extraction of shift pattern feature points, the above-mentioned weight factor is added to the traditional Euclidean distance-based clustering algorithm calculation formula. This allows the optimal shift point under multiple operating conditions to be considered simultaneously, achieving a statistically significant hybrid optimal shift pattern design. Attached Figure Description

[0053] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0054] Figure 1 This is a schematic diagram of the optimal shifting law design method of the present invention;

[0055] Figure 2 This diagram illustrates the optimal shift points and their weighting factors for two different operating conditions. The blue and red dots in the diagram represent the optimal shift points generated under the two operating conditions (urban and suburban conditions), respectively.

[0056] Figure 3 A schematic diagram of the clustering extraction process considering the weighting factors for two operating conditions. Detailed Implementation

[0057] Numerous specific details are set forth in the following description to provide a full understanding of the invention. However, the invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0058] The terminology used in one or more embodiments of this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of this specification. The singular forms “a,” “described,” and “the” as used in one or more embodiments of this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0059] It should be understood that although the terms first, second, etc., may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first may also be referred to as second without departing from the scope of one or more embodiments of this specification, and similarly, second may also be referred to as first. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0060] The present invention will now be described in further detail with reference to the accompanying drawings:

[0061] A method for optimizing hybrid gear shifting patterns based on operating condition data includes the following steps:

[0062] Obtain typical operating condition information for the target vehicle model;

[0063] Based on typical operating condition information and the established value function, the optimal control trajectory is generated.

[0064] Determine the globally optimal control law based on the optimal control trajectory;

[0065] Based on the global optimal control law, the shift point is obtained in the coordinate system;

[0066] The shift points are clustered using the K-means clustering algorithm to obtain cluster centers.

[0067] The shift curve is generated by fitting the cluster centers;

[0068] Based on the shift curve, the optimal shift point is determined and the shift operation is executed.

[0069] Preferably, based on typical operating condition information and the established value function, the optimal control trajectory is generated, specifically including the following steps:

[0070] Establish a value function based on the state variables and control variables of typical operating conditions;

[0071] Based on the value function, the optimal control trajectory is generated using a dynamic programming algorithm.

[0072] Preferably, the formula for the value function is:

[0073]

[0074] In the formula: N is the total number of steps in the discrete driving condition; L is the instantaneous cost function; Ω is the adjustable penalty factor for shift frequency; J is the value function; x k It is a state variable; u k It is a control variable; P bat It refers to battery power.

[0075] Preferably, the state variables include the battery state of charge (SOC) and the gearbox gear.

[0076] The control variables include the requested torque Trq and the transmission shift command shift;

[0077] The formulas for state variables and control variables are:

[0078]

[0079] Preferably, the formula for calculating the optimal control trajectory is as follows:

[0080]

[0081] Preferably, the shift point is obtained in the coordinate system according to the global optimal control law, specifically including the following steps:

[0082] According to the global optimal control law, the shift point is obtained in the corresponding vehicle speed-pedal opening or vehicle speed-pedal opening-gradient coordinate system.

[0083] Preferably, the shift points are clustered using the K-means clustering algorithm to obtain cluster centers, specifically including the following steps:

[0084] Determine the number of cluster centers, k;

[0085] Randomly select k points as initial cluster centers, representing k clusters;

[0086] Calculate the Eulerian distance between each shift point and each initial cluster center, and assign each shift point to the cluster with the closest Eulerian distance;

[0087] Based on the shift point of each cluster, recalculate the new cluster centers;

[0088] This continues until the deviation between the previous cluster center and the current cluster center is less than the deviation threshold.

[0089] Preferably, the formula for calculating the weighting function is:

[0090]

[0091] In the formula: f v (x) is an sigmoid function. It is a combined Gaussian function, which represents a weighted function of the proportions of speed and pedal position at a certain working condition; λ, a k b k β is an adjustment parameter;

[0092] The formula for calculating the Euler distance is:

[0093]

[0094] In the formula: D(x, μ) is the cluster distance considering weights.

[0095] Preferably, the formula for recalculating the new cluster centers is:

[0096]

[0097] Where: μ′ j It is a new cluster center.

[0098] To better illustrate the technical effects of the present invention, the present invention provides the following specific embodiments to illustrate the above technical process:

[0099] Example 1: A method for optimizing hybrid shifting patterns based on operating condition data, such as... Figure 1 As shown, it includes:

[0100] The present invention first requires the collection of typical operating condition information of the target vehicle model. This operating condition information can come from the operating conditions specified in the national standard, or it can be road spectrum information actually collected in the application scenario.

[0101] After obtaining typical operating condition information, offline global optimization calculations will be performed to determine the optimal shifting time based on this condition. First, appropriate state variables are selected to construct the state equations. Here, we take an electric vehicle as an example. Since typical operating condition information, including vehicle speed and gradient, is known, we only need to use the battery state of charge (SOC) and the gearbox gear as state variables.

[0102] Since the optimization objective is to determine the optimal shift point under specified operating conditions, to simplify the control variables, only the requested torque Trq and the transmission shift command shift can be selected. The equations for the system's state variables and control variables can be described as follows:

[0103]

[0104] For this discrete optimization problem, considering the requirement for smooth power delivery during transmission control, its value function should take into account not only system energy consumption but also the suppression of cyclic shifting. Therefore, the value function can be expressed as follows, mainly including battery power consumption and shift penalty.

[0105]

[0106] In the formula: N is the total number of steps in the discrete driving condition; L is the instantaneous cost function; Ω is the adjustable penalty factor for shift frequency; J is the value function; x k It is a state variable; u k It is a control variable; P bat It refers to battery power.

[0107] For this problem, a dynamic programming (DP) algorithm based on the Bellman optimization principle is used for solution. This involves calculating and solving the cost function values ​​for all possible states at all stages, starting from the last step. Therefore, the global optimization strategy problem can be divided into a series of secondary recursive equations. The optimal transition cost for each state is solved by working backward from the final stage, while adding necessary penalty costs to unsatisfactory final states. The solution is then applied forward to determine the optimal control trajectory, as shown in the following equation:

[0108]

[0109] In the formula: This represents the minimum cost of each state transition. This represents the optimal control law at each step.

[0110] It should be noted that when using DP to solve for the optimal shift point for pure electric vehicles, the solution is different from that for general hybrid electric vehicles (HEVs). Electric vehicles are not sensitive to the final state of charge (SOC), and the final state of the gear does not need to be set with special requirements. Therefore, their final state is generally unconstrained. Except for some unreachable state variables, no penalty cost needs to be set for the other states.

[0111] On the other hand, since variables such as requested torque Trq and battery state of charge SOC are continuous variables, there may be situations where the state transition at stage k cannot fall on the existing state grid. In this case, the state transition cost from stage k+1 to the final stage is approximately determined using Lagrange interpolation, and then the corresponding state transition cost at stage k is calculated.

[0112] To obtain the globally optimal control law [shift1, shiff2, shift3... shift] N After this, the optimal shift point can be obtained in the corresponding coordinate system such as vehicle speed-pedal opening (two parameters) or vehicle speed-pedal opening-gradient (three parameters). The correlation between these shift points is poor, and it is generally not possible to directly generate the shift curve (surface) required to construct the shift pattern through function fitting. Therefore, it is necessary to further solve for the shift pattern feature points (optimal shift points). These feature points will be used in the transmission controller (TCU) as an online available shift strategy.

[0113] This invention employs the K-means clustering algorithm, which boasts high computational speed and good convergence, to extract feature points of shifting patterns. If based on an n-parameter shifting pattern, it suffices to project the optimal shifting point onto the space constructed by the considered n parameters; the corresponding point can be described as...

[0114] Since subsequent calculations are mainly based on spatial Eulerian distance, and different parameters have different units, the above points are first normalized to unify them into dimensionless units.

[0115] Set k random cluster centers as The core of the K-means clustering algorithm is to find the cluster center that is closest to each point by Euclidean distance and group them into a cluster (class), thus minimizing the squared error between the cluster center and the points in the corresponding cluster.

[0116]

[0117] The above formula defines the nearest cluster center μ for each point in the dataset. j and its cluster c (j) Since the initial cluster centers are randomly given, they need to be updated. Generally, the geometric center of the data in each cluster is calculated as the new cluster center.

[0118]

[0119]

[0120] The ultimate goal of K-means clustering is to minimize the sum of squared errors across all k clusters, as shown in the equation. It should be noted that solving this value function is an algorithmic nondeterministic polynomial problem. Simply put, even for k=2, the solution can only converge to a local optimum, and a globally optimal result cannot be guaranteed. Therefore, in practical applications, iterative trial and error must be performed considering the characteristics of the data. This limitation also makes its efficiency in obtaining effective information in offline applications relatively low.

[0121] The K-means algorithm requires determining the number of clusters initially; otherwise, it needs to iterate repeatedly based on the results. To speed up the selection process, the Analytic Hierarchy Process (AHP) is used to search for a suitable number of cluster centers, k.

[0122] The Hierarchical Analysis (AAMI) is a structured method for analyzing complex decisions based on mathematical theory. It can also be viewed as a clustering method that merges the two closest nodes and repeats the clustering process until only one cluster remains. In other words, this method can only operate from the initial points and excels at defining the optimal number of clusters. Since Euclidean distance is the sole criterion for determining the number of nodes, a dendrogram can be generated to describe the hierarchical binary clustering tree, where the height of each U-shaped line represents the distance between the two connected nodes. The two closest nodes are merged sequentially until only one cluster remains.

[0123] The threshold is typically set to 15% of the maximum distance shown to assess the level of data divergence and determine the number of cluster centers.

[0124] After determining the number of clusters, the K-means clustering algorithm can divide the data into different groups based on the distance of normalized features. The conventional algorithm consists of the following steps:

[0125] Step (1): Randomly place the determined k points into the space represented by the objects used as the initial cluster centroids for clustering.

[0126] Step (2): Calculate the distance between the data point (optimal shift point) and the cluster center, and then assign the point to the nearest center.

[0127] Step (3): After all objects are assigned, for each cluster, recalculate the new average position of the k centers based on the selected data points in the same group, such as... Figure 3 As shown.

[0128] Step (4): Repeat steps (2) and (3) until the deviation between the current center and the new center is less than a certain value (deviation threshold s) to reasonably group the objects.

[0129] In step (2) above, Euclidean distance is typically used to calculate the distance between each point. Considering the different characteristics of the data, a weighting factor that proportionally affects the distance calculation results can be used to combine data from various driving scenarios.

[0130] The weighting factor can be calculated directly by proportionally dividing the usage frequency of each operating condition, for example, urban driving conditions: highway driving conditions: mountain driving conditions = 5:4:1. This allows for the direct acquisition of a set of weighting factors to adjust the relative positions of shift pattern feature points and optimal shift points under different operating conditions. Alternatively, based on the frequency distribution of parameters such as pedal opening and vehicle speed, a functional relationship can be fitted to determine the weighting factors for each dimension in distance calculation, such as... Figure 2 As shown.

[0131] As mentioned above, the contributions of each point vary greatly depending on the distribution. Since the data is described in two dimensions, weighting functions for speed and pedal position should be determined separately. To reasonably capture the weighting characteristics of urban and suburban driving conditions, different types of S-functions are used, as shown in the equation.

[0132]

[0133] In the formula: fv(x) is an S-shaped function, fp(x) is a combined Gaussian function, which respectively represent the weighted functions of the proportion of speed and pedal position at a certain working condition midpoint, and the remaining variables are adjustment parameters that can change the trend of the relationship.

[0134] By using the weighting factor shown in the formula, overlapping points can avoid interfering with each other after mixing shift points from different cases. Furthermore, due to the weighted contribution, it not only appropriately considers all features but also avoids the balancing effect between the two sets of data. It should be noted that the distance calculation in step (2) should be replaced with:

[0135]

[0136] In the formula: D(x, μ) is the cluster distance considering weights.

[0137]

[0138] Where: μ′ j It is a new cluster center.

[0139] In the several embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules, units, or units is merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units, modules, or components may be combined or integrated into another device, or some features may be ignored or not executed.

[0140] The units may or may not be physically separate. The components shown as units can be one or more physical units, meaning they can be located in one place or distributed in multiple different locations. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0141] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0142] In particular, according to embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the functions defined in the methods of this invention. It should be noted that the computer-readable medium described above in this invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof.

[0143] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0144] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions within the technical scope disclosed in the present invention should be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for optimizing hybrid gear shifting patterns based on operating condition data, characterized in that, Includes the following steps: Obtain typical operating condition information for the target vehicle model; Establish a value function based on the state variables and control variables of typical operating conditions; Based on the value function, the optimal control trajectory is generated using a dynamic programming algorithm. The formula for the value function is: In the formula: N is the total number of steps for discrete driving conditions; L is the instantaneous cost function; It is an adjustable penalty factor for shift frequency; It is a value function; It is a state variable; It is a control variable; It refers to battery power; The state variables include the battery state of charge (SOC) and the gearbox gear. The control variables include the requested torque Trq and the transmission shift command shift; The formulas for state variables and control variables are: Determine the globally optimal control law based on the optimal control trajectory; Based on the global optimal control law, the shift point is obtained in the coordinate system; The shift points are clustered using the K-means clustering algorithm to obtain cluster centers. The shift curve is generated by fitting the cluster centers; Based on the shift curve, the optimal shift point is determined and the shift operation is executed.

2. The method for optimizing hybrid shifting patterns based on operating condition data according to claim 1, characterized in that, Based on the global optimal control law, the shift point is obtained in the coordinate system, specifically including the following steps: According to the global optimal control law, the shift point is obtained in the corresponding vehicle speed-pedal opening or vehicle speed-pedal opening-gradient coordinate system.

3. The method for optimizing hybrid shifting patterns based on operating condition data according to claim 1, characterized in that, The shift points are clustered using the K-means clustering algorithm to obtain cluster centers. The specific steps include: Determine the number of cluster centers, k; Randomly select k points as initial cluster centers, representing k clusters; Calculate the Eulerian distance between each shift point and each initial cluster center, and assign each shift point to the cluster with the closest Eulerian distance; Based on the shift point of each cluster, recalculate the new cluster centers; This continues until the deviation between the previous cluster center and the current cluster center is less than the deviation threshold.

Images