PSO-based optimization method for simulation data of vehicle transmission system torsional vibration characteristics

Abstract

The present invention proposes a method for optimizing simulation data of vehicle transmission system torsional vibration characteristics based on particle swarm optimization (PSO), comprising: step S1: constructing a simulation model of the transmission system's torsional vibration characteristics in torsional vibration characteristic simulation software based on complete vehicle simulation information; step S2: using the simulation software to simulate and analyze the vehicle transmission system and extracting the second-order angular acceleration data of the transmission input shaft after the power source vibration is damped by the shock absorber; step S3: optimizing parameters during the calculation of the RMS value of the transmission input shaft's second-order angular acceleration using a particle swarm optimization method; and step S4: calculating the RMS value of the transmission input shaft's second-order angular acceleration using the parameters obtained through the particle swarm optimization. The present invention optimizes the number of points selected and repeatedly calculated in the RMS, resulting in an RMS value that best reflects the trend changes of the original data.

Description

Technical Field

[0001] The present invention belongs to the field of simulation of vehicle transmission, and in particular relates to a method for optimizing simulation data of torsional vibration characteristics of a vehicle transmission system based on PSO. Background Art

[0002] With the current domestic and international automotive markets continuing to expand and broaden, more and more vehicles are entering the market. Before a vehicle is delivered to its user, its ride comfort must be fully considered, which is directly related to the vehicle's vibration suppression performance. Generally speaking, the main source of vehicle vibration is irregular excitation from the engine or other power source, which propagates throughout the drivetrain, reaching the body, seats, and other locations. The torsional vibration damper (or dual-mass flywheel, etc.) plays a key role in reducing torsional vibration in the drivetrain.

[0003] Traditional vehicle transmission test benches are subject to high costs, complex construction, and long development cycles. Parts companies can verify the compatibility of their developed vibration reduction products with vehicle models using simulation software related to the torsional vibration characteristics of vehicle transmission systems. This allows them to provide solutions for the parts required by vehicle manufacturers, thereby improving product compatibility and shortening new vehicle development cycles. OEMs and parts manufacturers are particularly concerned with the RMS value of the second-order angular acceleration of the transmission input shaft, which is the result of the source engine's irregular vibration being damped by the shock absorber. Therefore, optimizing the relevant parameters in the RMS calculation process is particularly important in order to ensure that the RMS value better reflects the size and trend of the original data. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for optimizing simulation data of torsional vibration characteristics of vehicle transmission system based on PSO, and to solve the RMS value of simulation data obtained by vehicle torsional vibration characteristics simulation software. The RMS is calculated by calculating one RMS for every few data points, rather than a parameter optimization method for all data points.

[0005] Vehicle manufacturers and parts manufacturers are particularly interested in the RMS value of the second-order angular acceleration of the transmission input shaft in the vehicle's transmission system. This value is used to characterize the damping effect of the torsional vibration damper on the source engine vibration. Therefore, it is necessary to export the numerical values ​​obtained from the simulation software and then perform post-processing with MATLAB to calculate the RMS value. This invention uses a particle swarm optimization algorithm to optimize the number of points selected for RMS value calculation and the number of points for repeated calculation. The RMS calculated from the optimized parameters can better reflect the main characteristics of the original data. The technical solution adopted is as follows:

[0006] A method for optimizing simulation data of torsional vibration characteristics of a vehicle transmission system based on PSO includes the following steps:

[0007] Step S1: Building a simulation model of the torsional vibration characteristics of the transmission system in the torsional vibration characteristics simulation software based on the simulation information of the entire vehicle. Specifically:

[0008] Step S1.1: Based on the transmission system topology diagram of the entire vehicle, the entire vehicle transmission system torsional vibration characteristics overall framework modeling is performed using the lumped mass method in the simulation software.

[0009] Step S1.2: Set the corresponding parameters in the model according to the geometric shape, torsional stiffness, rotational inertia, torsional damping and gear ratio of each component of the vehicle transmission system.

[0010] Step S1.3: Build a steady-state torque output module at each speed according to the vehicle power source external characteristic curve.

[0011] Step S2: Use simulation software to perform simulation analysis on the vehicle transmission system and extract the second-order angular acceleration data of the transmission input shaft after the power source vibration is damped by the shock absorber. Specifically:

[0012] Step S2.1: Perform simulation using the vehicle transmission system torsional vibration simulation model built in step S1 based on the simulation software.

[0013] Step S2.2: Then, perform order analysis and use Fast Fourier Transform (FFT) to extract the second-order angular acceleration of the transmission input shaft, which is of most concern to manufacturers.

[0014] Step S3: Optimizing the parameters in the process of calculating the RMS value of the second-order angular acceleration of the transmission input shaft using the particle swarm optimization method. Specifically:

[0015] Step S3.1: Preprocess the original data and group the original data according to the number of groups calculated by the RMS value.

[0016] Step S3.2: The original value (the pre-processed second-order angular acceleration of the transmission input shaft) is used as the comparison value. The algebraic sum of the squares of the difference between the RMS value and the comparison value is used as the particle fitness. A particle swarm optimization algorithm is used to optimize the number of points used for RMS calculation and the number of points for repeated calculations, aiming to minimize the penalty for particle fitness.

[0017] Step S4: Calculate the RMS value of the second-order angular acceleration of the transmission input shaft using the parameters obtained by particle swarm optimization: the number of selected points for RMS value calculation and the number of repeated calculation points.

[0018] Compared with the prior art, the advantages of the present invention are:

[0019] The optimization of parameters in the RMS calculation of vehicle driveline torsional vibration simulation software is carried out. The optimization of the number of points selected and repeated calculations in the RMS calculation is carried out to ensure that the RMS value can best reflect the trend changes of the original data. Furthermore, the parameter optimization method using particle swarm optimization can save time and avoid manual parameter tuning, which wastes time and causes data characteristics to deviate significantly from the original data. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 for Figure 1 Flowchart of parameter optimization technology based on PSO;

[0021] Figure 2 It is the topological diagram of the vehicle transmission system structure;

[0022] Figure 3 The vehicle transmission system torsional vibration simulation lumped mass method model built based on the transmission system torsional vibration characteristics simulation software in the present invention;

[0023] Figure 4 This is the amplitude-frequency diagram of the second-order angular acceleration order analysis in the angular acceleration order analysis of the transmission input shaft;

[0024] Figure 5 is the original data diagram of the angular acceleration of the transmission input shaft;

[0025] Figure 6 This is the result diagram of RMS solution of simulation parameters using PSO tuning parameters. DETAILED DESCRIPTION

[0026] The following, with reference to a schematic diagram, provides a more detailed description of the PSO-based vehicle transmission system torsional vibration characteristics simulation data optimization method of the present invention. This diagram illustrates a preferred embodiment of the present invention. It should be understood that those skilled in the art may modify the invention described herein while still achieving its advantageous effects. Therefore, the following description should be understood as a general guide for those skilled in the art and not as a limitation of the present invention.

[0027] A PSO-based optimization method for vehicle transmission system torsional vibration characteristics simulation data, such as Figure 1 As shown, specifically including:

[0028] Step S1: Based on the simulation information of the whole vehicle, a simulation model of the torsional vibration characteristics of the transmission system is constructed in the simulation software. Specifically:

[0029] Step S1.1: The transmission system topology diagram of the vehicle is simulated in the simulation software using the lumped mass method. Figure 2 The vehicle transmission system structural topology diagram shown is used to model the torsional vibration characteristics of the entire vehicle transmission system.

[0030] Design a transmission system model that includes mass points and connection elements, clarify the types of connection elements, and their associated physical quantities, such as possible parameters such as stiffness, damping, transmission ratio, etc.

[0031] Step S1.2: Set the corresponding parameters in the model according to the torsional stiffness, rotational inertia, torsional damping and gear ratio of each transmission component designed for the vehicle transmission system.

[0032] Among them, the concentrated mass method should be used to convert the translational inertia of the vehicle body into rotational inertia. According to the law of conservation of energy, its movement is converted into rotation with the wheel radius as the radius, and the inertia force of the vehicle body is converted into inertia moment with the lever arm as the wheel radius.

[0033] Step S1.3: Build a steady-state torque output module at each speed according to the vehicle power source external characteristic curve.

[0034] Power modules can be flexibly added or removed for single or multiple power sources of gasoline and diesel engine vehicles, pure electric vehicles, and hybrid vehicles.

[0035] Step S2: Use simulation software to perform simulation analysis on the automobile transmission system, and extract the second-order angular acceleration data of the transmission input shaft after the vibration reduction treatment by the torsional vibration damper (or dual-mass flywheel, etc.).

[0036] Step S2.1: Using the Figure 3 , simulation is performed based on the vehicle transmission system torsional vibration simulation model built by simulation software.

[0037] If it is a pure electric vehicle, set the opening to 1; if it is a fuel vehicle, set the opening to 1; if it is a hybrid vehicle, consider the special working condition that when the throttle opening is 1, set the opening according to the external characteristics of the drive motor so that the torque output to the transmission system by the fuel engine and the drive motor is equal.

[0038] In addition to the three common typical types of vehicles mentioned above, the settings of other types of vehicles are also considered in the same way. The load setting can be added after the body module.

[0039] Step S2.2: Use the fast Fourier transform method to process the vibration angular acceleration data, calculate the vibration order amplitude-frequency characteristics, and then obtain the RMS value for evaluating the vibration characteristics. In this invention, only the second-order angular acceleration of the transmission input shaft is concerned, see Figures 4-5 .

[0040] Figure 4 In the figure, the horizontal axis is the frequency and the vertical axis is the speed of the transmission input shaft;

[0041] Figure 5 In , the horizontal axis is the transmission input shaft speed, and the vertical axis is the transmission input shaft angular acceleration;

[0042] "2" means the second-order angular acceleration of the transmission input shaft.

[0043] Step S3: Optimizing the parameters for calculating the RMS value of the second-order angular acceleration of the transmission input shaft using a particle swarm optimization method.

[0044] Step S3.1: Pre-processing the original data, ie, the value of the second-order angular acceleration of the transmission input shaft (data of the second-order angular acceleration in the speed domain).

[0045] The original data is approximately considered to obey the normal distribution, so the outliers are eliminated based on the 3σ principle. 99.73% of the data will fall within u±3σ, that is, the probability that the data is distributed in (u-3σ,u+3σ) is 0.9973.

[0046] The removed abnormal points are replaced with the corresponding values ​​of the previous speed value calculated by the simulation software, see Figure 5 .

[0047]

[0048] -Second-order angular acceleration of the transmission input shaft.

[0049] Group the original data into groups with the same number of groups as those used to calculate the RMS value.

[0050] If there are any remaining numerical points after grouping, they should be the second-order angular acceleration corresponding to the highest speed, which is not very important for evaluating the vibration reduction effect of the shock absorber. Therefore, they are not considered.

[0051]

[0052] x represents the number of groups for calculating the RMS value, m represents the number of original data points obtained by the simulation software, k represents the number of repeated calculation points for calculating the RMS value (referring to the number of repeated calculation points between two adjacent groups), and n is the number of values ​​in each group.

[0053] Step S3.2: The original value, i.e., the second-order angular acceleration of the transmission input shaft, is pre-processed and used as the comparison value. Because the original data is divided into multiple groups for RMS value calculation, the RMS value of each group of data is calculated.

[0054]

[0055] In formula (3), i is each group, RMS i The RMS value calculated for each data group, n i The number of data in each data group, d ij is the value of the jth data point in the i-th data group.

[0056] The mean for each group is calculated as follows:

[0057]

[0058] The particle fitness is penalized at the cost of the square algebraic sum of the difference between each group's RMS calculation value and the average of each group's corresponding comparison value. As shown in formula (5):

[0059]

[0060] In formula (5), J is the fitness of the particle, d i The mean value for each data group, from 1 to x.

[0061] Using the particle swarm optimization algorithm, the initial value of the particle is given artificially, the initial position of the particle is [n, k], the initial velocity direction of the particle is [v n ,v k ], that is, each particle contains two-dimensional information.

[0062] The two parameters of the number of selected points and the number of repeated calculation points for RMS value calculation are optimized in order to minimize the cost penalty.

[0063] The particle's position and velocity are updated at each iteration as follows:

[0064]

[0065] In formula (6):

[0066] is the velocity of particle i at time t, is the velocity of particle i at time t+1, is the position of particle i at time t, is the position of particle i at time t+1;

[0067] p i is the historical optimal position of particle i, g is the global optimal position, w is the inertia weight, c1, c2 are learning factors, and r1, r2 are random numbers.

[0068] When the set iteration error condition is met or the maximum number of iterations is reached, the optimal parameter combination [n opt ,k opt ].

[0069] Step S4: Optimize the parameters obtained by particle swarm optimization [n opt ,k opt ], Matlab was used to calculate the RMS value of the second-order angular acceleration of the transmission input shaft. Finally, a fitting curve was obtained through cubic spline curve fitting.

[0070] The above description is merely a preferred embodiment of the present invention and does not limit the present invention in any way. Any person skilled in the art who, without departing from the scope of the present invention, makes any equivalent substitution, modification, or other changes to the technical solution and technical content disclosed in the present invention shall be deemed to be within the scope of the present invention and still fall within the scope of protection of the present invention.

Claims

1. A method for optimizing simulation data of torsional vibration characteristics of vehicle transmission system based on PSO, characterized in that: The following steps are involved: Step S1: constructing a simulation model of the torsional vibration characteristics of the transmission system in torsional vibration characteristics simulation software based on the simulation information of the entire vehicle; Step S2: using simulation software to perform simulation analysis on the vehicle transmission system, and extracting the second-order angular acceleration data of the transmission input shaft after the power source vibration is damped by the shock absorber; Step S3: Optimizing parameters in the process of calculating the RMS value of the second-order angular acceleration of the transmission input shaft using a particle swarm optimization method; Step S4: Calculating the RMS value of the second-order angular acceleration of the transmission input shaft using the parameters obtained by particle swarm optimization; Step S3 includes the following steps: Group the original data into groups with the same number of groups as those used to calculate the RMS value; ； Indicates the number of groups for calculating RMS values, Indicates the number of original data points obtained by the simulation software, Calculate the RMS value and repeat the calculation points. is the number of values ​​in each group; The original value, i.e., the second-order angular acceleration of the transmission input shaft, is used as the comparison value after numerical preprocessing. Since the original data is divided into multiple groups to obtain the RMS value, the RMS value of each group of data is calculated; ； For each group, Calculated for each data group value, The number of data in each data group, For the The first data group data point values; The mean for each group is calculated as follows: ； The particle fitness is penalized at the cost of the square algebraic sum of the differences between each group's RMS calculation value and the average of each group's corresponding comparison values; ； is the fitness of the particle, is the mean value for each data group, The value ranges from 1 to ; Using the particle swarm optimization algorithm, the initial value of the particle is given artificially, and the initial position of the particle is , the direction of the particle's initial velocity , that is, each particle contains two-dimensional information; Optimize the two parameters of the number of selected points and the number of repeated calculation points for RMS value calculation in order to minimize the cost penalty; When the set iteration error condition is met or the maximum number of iterations is reached, the optimal parameter combination is output .

2. The method for optimizing simulation data of vehicle transmission system torsional vibration characteristics based on PSO according to claim 1, characterized in that: Step S1 specifically includes the following steps: Step S1.1: Based on the transmission system topology diagram of the entire vehicle, the entire vehicle transmission system torsional vibration characteristics overall framework modeling is performed using the lumped mass method in the simulation software; Step S1.2, setting the corresponding parameters in the model; Step S1.3: Build a steady-state torque output module at each speed according to the vehicle power source external characteristic curve.

3. The method for optimizing simulation data of vehicle transmission system torsional vibration characteristics based on PSO according to claim 1, characterized in that: Step S2 specifically includes the following steps: Step S2.1: Simulate the vehicle transmission system torsional vibration simulation model built using simulation software; Step S2.2: Perform order analysis to extract the second-order angular acceleration of the transmission input shaft.

Images