1Crawler vehicle ride simulation model establishment
1. 1 tracked vehicle virtual prototype model
In the research of vehicle ride comfort simulation in the literature, the rigid parts model is usually used for every part of the vehicle. Taking into account that the actual vehicle running process, parts will produce flexible deformation, so the study will be the body as a flexible body simulation calculations to reduce the simulation error.
The process of establishing the virtual prototype model of tracked vehicles is as follows: (1) Using three-dimensional CAD software to establish the model of each part of the vehicle; (2) Extracting the midplane of the vehicle body and introducing it into the finite element analysis software. According to the actual structure of the vehicle, add supporting beam units to the vehicle. The body is divided into grids to build a flexible car body; (3) Each component model and flexible car body model are imported into multi-body dynamics software, and other parts models such as balance elbow, road wheel, track, and counterweight are added; (4) Based on actual vehicle data, the parameters of the system such as suspension, seat, track shoes, and ground were set, and the corresponding constraints and force relationships were added to establish a rigid-flexible hybrid virtual prototype model of the tracked vehicle.
1. 2 random road model
The unevenness of the driving ground is the main cause of vehicle vibration. The research shows that the road surface irregularities have the characteristics of randomness, stability and ergodicity, which can be described by the stochastic stochastic process theory. The line of intersection between the vertical profile of the road and the surface of the road is usually used as a sample of road surface roughness. The power spectrum density of the sample is used to describe the road surface. The power spectral density is the most important mathematical feature of road surface roughness. It can represent the distribution of energy of the road surface irregularity in the spatial frequency domain to explain the road surface roughness or the structure of the road surface wave.
In general, the height profile of the road profile follows the Gaussian distribution. For smooth Gaussian stochastic processes there are several ways to generate time-domain models of road roughness. The main methods are: filtered white noise generation method, discrete-time random sequence generation method based on rational function PSD model, harmonic superposition method, fast Fourier inverse transform generation method based on power function power spectrum, and so on. Harmonic superposition method is used in this paper to construct a time domain model of random road roughness and obtain the contour curve of a specific road surface.
The principle of harmonic superposition method to fit uneven pavement is to set a smooth, ergonomic Gaussian process with an average value of 0. You can use different forms of trigonometric series to perform simulations. This article uses a finite number of discrete spatial frequencies. The trigonometric series describes the random process q( l )= N k = 1 a sin(2 n K l+ K) (1) where l denotes the path length and K denotes the phase angle, which is uniformly distributed in the 0 2 interval. Random variable; N represents the reference spatial center frequency, N = 3 (lgn u - lgn l) / lg2; nl represents the lower bound spatial frequency of the road spectrum, nl = 0. 011 m - 1; nu represents the upper limit spatial frequency of the road spectrum , nu = 6. 667 m - 1; nK denotes the Kth center frequency of the reference space, nk = 2 (K - 0.5)nl; aK denotes the amplitude of the discrete spatial frequency trigonometric series, a K 0 . 51 n 0 (G qn 0 / nl) 0.5 (2 - K / 6); n 0 denotes the reference spatial frequency, n 0 = 0. 1 m - 1; G q (n 0) denotes the road surface roughness coefficient.
The international standard ISO2631 classifies the road surface into 8 levels according to the power spectral density. The level of G q (n0) from level A to level H increases step by step, indicating that the unevenness of the road surface () becomes correspondingly larger, and it stipulates that each level of road surface The range and mean value of the roughness coefficient. According to the corresponding G q (n0) in the road surface level selection criteria to be generated, the value of the road surface roughness random process q( l ) is calculated. The real vehicle test road surface studied in this paper is a cobblestone road surface. The F-grade road surface is used for simulation. The F-level road surface roughness coefficient is in the range of 8 192 10 - 6 32 768 10 - 6 m 3 and the simulated road surface is G q (n 0 ) = 10 000 10 - 6 m 3, 200 m long F-grade road surface is produced, and the profile of the pavement profile is shown in 1. 1 The horizontal axis represents the length and the vertical axis represents the road surface height.
After generating the road surface roughness data, the pavement profile file is prepared according to the triangular unit method and the multi-body dynamics software required format. In the multi-body dynamics software environment, the F-level road surface model is generated.
2 Driving Ride Evaluation Method
According to ISO/ISI 2631-1:1997(E), the vibration and shock of a human body is subjected to the whole-body vibration evaluation standard. When evaluating the vibration, the total vibration acceleration is used in a human sitting vibration model with a basic frequency range of 0.50 Hz. The values ​​of square roots and human subjective feelings are used to judge the occupant comfort and the rider comfort is used to evaluate the running smoothness of the vehicle. The standard gives a detailed frequency weighting function and axis weighting factor based on the test, as well as a clear comfort limit. The specific method for calculating the RMS acceleration is as follows: First, perform frequency spectrum analysis on each axial acceleration time course a (t) to obtain the power spectral density function Ga(f). Then, according to equation (2), you can calculate the frequency weighted value. The root mean square value of the acceleration a waw = 0. 5 (2), where w (f) is the frequency weighting function; f represents the vibration frequency, for the vertical direction (z axis), w (f) = 0.50. 50.5 (5) According to equation (5), the weighted acceleration rms value at the driver's seat position can be calculated to evaluate the running smoothness of the vehicle.
3 Ride Simulation and Analysis
3.1 Simulation and Data Processing Process
Based on the tracked vehicle virtual prototype and the random road model established, the tracked vehicle ride comfort simulation was performed according to the ride comfort evaluation method. The software environments used were: three-dimensional modeling software Pro/E, finite element analysis software ANSYS, multi-body dynamics software RecurDyn, and data processing using MATLAB. The entire simulation process is shown as 2.
Some of the simulation parameters in ride simulation are set as follows: vehicle travel speed is: 20/30/40 (km/h); simulation time is: 35/30/25(s); simulation road length: 200(m); Ground contact stiffness: 8 000 (N/m); Track and ground contact damping coefficient: 10 (Ns/m); Pavement level: F.
3.2 Ride Simulation
According to the simulation parameter settings, tracked vehicles were simulated at speeds of 20, 30, and 40 km/h on the F-class road surface. Acquire the vibration acceleration curve of the driver's seat at each speed, and calculate the RMS of the vibration acceleration at each speed using equations (2) and (5).
Shown as the vibration acceleration curve of the driver's seat at the speed of 20 km/h (a), vertical direction (b) and left and right direction (c).
It can be seen from the above that the 0 5 s vehicle experienced the process from being suspended to landing and then accelerating to the set speed. In this process, the vibration accelerations in the three directions are relatively large, and there are great fluctuations. From the 5. 0 s to the end of the simulation, the vehicle was driving on an uneven F-level road surface, and the vibration acceleration curve began to appear irregular fluctuations. This was reflected in the uneven vibration of the vehicle.
The three directions of vibration acceleration curve data in the driver's seat are converted into data files for data processing, and the calculated total root mean square values ​​of the acceleration in the three directions of the vehicle can be calculated.
In the same way, the vibration acceleration curves of the driver's seat at 30 km/h and 40 km/h are obtained by simulation in this paper respectively. After weighted calculation, the total acceleration RMS value is obtained to evaluate the vehicle's Smooth ride.
3.3 Comparison of simulation results with actual vehicle test data
In order to verify the credibility of the virtual prototyping model, this study calculated and analyzed the vibration acceleration test data of the real vehicle walking test and compared it with the computer simulation results.
The test equipment adopts Wavebook/516A, a portable high-speed data acquisition and analysis system from American IOtech. The system can measure a variety of parameters, support multiple data acquisition and signal analysis software, and can complete multi-parameter signal acquisition. The test was carried out on a cobblestone road. The sampling frequency of the equipment was 51 200 Hz. The sampling time was 25 s. At the vehicle speeds of 20, 30, and 40 km/h, the vehicle's seat, gearbox, and engine were collected separately. Vibration acceleration data of the part.
In order to perform occupant comfort calculations, the driver seat channel test data is extracted, and the data is re-sampled (every 50 data samples are taken once), and a data file is generated to calculate the rms acceleration rms value. Similarly, for the computer simulation model system, vibration acceleration data of the same position is collected after simulation. Then calculate the total acceleration rms value for the two kinds of data to evaluate the occupant comfort. In order to compare the two calculation results, according to the calculation formula: (aw simulation-aw measurement) / aw measurement, calculate the difference. At speeds of 20, 30, and 40 km/h, respectively, the difference between the simulated and measured data was: 1.33%, 6.8%, and 9.88%.
From the calculation results, it can be seen that the simulation error is generally within an acceptable range. Especially at low speed, the results of computer simulation and actual vehicle test are very close, which shows that the ride comfort simulation model system established in the paper is reasonable. It also shows that for a complex mechanical system of tracked vehicles, performance analysis and evaluation through virtual prototyping is feasible, and it can be used to predict and evaluate the ride comfort of vehicles, which evaluates the dynamic performance of tracked vehicles, as well as vehicle design and improvement. It has important practical significance.
After many simulation experiments and calculation results analysis, it is known that for the vehicle system parameters, actual or near-real data should be used, such as the stiffness and damping coefficient of the torsion bar spring and shock absorber, seat spring parameters, and track pre-tensioning force. The stiffness coefficient of the wheel, etc.; for the simulation of environmental parameters, should be consistent with the actual test environment, such as the road surface level and its parameters, track pads and ground contact parameters. The setting of these parameters has a great influence on the smoothness of the vehicle. Actual and reasonable parameters should be selected to reflect and evaluate the actual performance of the vehicle.
4 ends
By establishing a rigid-flexible hybrid virtual prototype model of a tracked vehicle and generating a ride comfort simulation system, the vehicle driving simulation is performed based on the F-grade random road model, and the simulation data is processed to analyze and evaluate the driving performance of the vehicle. After comparison and analysis with actual vehicle test data, it shows that the model and evaluation method established in the study are reasonable, and can be further extended to the simulation analysis and evaluation of the ride smoothness of various vehicles on various grades of road surface. Therefore, it can provide effective methods and theoretical references for the optimal design, performance analysis, and innovation of vehicles.
High Pressure Washing Machine,Car Washer,Washer Machine
High Pressure Cleaner ,Electric High Pressure Washer Co., Ltd. , http://www.china-pressurewasher.com