Research On 6-DOF Force Waveform Reproduction Control Method Of Spindle Coupled Road Simulator

The 6-DOF spindle coupled road simulator is the most representative equipment which is used to test vehicle durability in laboratory at the present stage. However, our country currently has no ability to independently develop such products, in which the research of control technology is the core issue. Based on the research project of “spindle coupled road simulation test bench” by Harbin Institute of Technology, this paper carries on the systematic analysis and research around these issues related to 6-DOF force time waveform reproduction control technology.Firstly, in this paper, the ADAMS model is modified to solve the divergence problem during simulation. Then a complete vehicle load model is established, and the vehicle simulation model of four angles configuration is given. Based on the established inverse kinematics model, Newton-Euler method was adopted to establish inverse dynamics models of hydraulic actuator, bell crank and bench by resonably using vector dot-product and cross-product to eliminate large amounts of the unknown constraining forces. So the model of the 6-DOF force of bench transformed into six hydraulic cylinder driving forces is obtained. After the comprehensive analysis of the load conditions, this paper establishes the two-degree of freedom hydraulic system model of the vertical direction and the single-degree of freedom hydraulic system model of the others. By using second-order oscillation section, inertia link, lag adjust and PID. etc, the ideal control effect of single channel is obtained.Because the setting of the excitation signal directly affects the precision of system identification, this paper firstly presented Welch method of time-domain signal transformed into power spectral density and the harmony superposition method and time domain randomization method of power spectral density transformed into time-domain signal. Two identification methods of parameter identification and nonparametric identification are proposed for system identification. Parameter identification adopts the single-point excitation of random signal. The model structrue is estimated based on residual sum of squares, and the SISO recursive least square algorithm is extended to the MIMO algorithm. Finally, the frequency response function matrix of the six-single degree of freedom is obtained. The H1 estimation method, H2 estimation method and the improved H3 and H4 estimation algorithm of nonparametric identification are given in turn. In order to improve identification precision, excitation signal uses multisine based on multi-point random excitation technique, and coherence function is considered as the criterion for the identification accuracy. Finally, this paper gets the law that identification accuracy of frequency response function is influenced by different sampling periods by simulation.As the system can not meet the control precision requirements only by servo control, this paper presents 6-DOF force waveform reproduction control technology. Fully considering actual road simulation test, the pavement roughness of the displacement and acceleration power spectral density are reproduced using the harmony superposition method, and takes it as the input of frequency response function identification and iterative control. The very small singular values of the frequency response function are processed with the singular value truncation method and the singular value modified method, and then the system impedance function is solved by using the generalized inverse theory, which will guarante the stability of the iterative algorithm. For the correction of the driving signal, piecewise iterative algorithm and variable correction coefficient algorithm are proposed to improve the convergence speed and stability of the iterative algorithm. Eventually, the overall system model of 6-DOF force waveform reproduction control is built. Through simulation, the ideal waveform reproduction effects of the six S-DOF force and the 6-DOF force are obtained. Finally, the new direct drive volume control electro-hydraulic servo system is used to carry out the experiment to verify the validity and universality of the time domain waveform reproduction algorithm.