Research On The Robust Optimization Design Of Disc Brake Systems With Interval-probability Uncertainty

With the development of the vehicle industry and car ownership increasing in the city, brake noise of vehicle disc brake system needs to be suppressed, because brake noise do harm to people's health. The analysis and optimization of the stability of vehicle disc brake systems is the most effective way to control the brake noise. The uncertain factors such as manufacturing error and wear exist in the vehicle disc brake systems inevitably, which result in the traditional analysis and optimization approaches unreliable. To suppress the brake noise of vehicle disc brake systems with uncertainty, an effective analysis and optimization method needs to be put forward.Supported by the independent project of State Key Laboratory of Advanced Automobile Body Design and Manufacture in Hunan University (Project approval number:71375004) and the Graduate Student Research Innovation Project of Hunan province (Project approval number:CX2013B143), this thesis carried out the stability analysis and the robustness optimization on vehicle disc brake systems with interval-probability uncertainty.The main research works and innovative achievements in the thesis are listed as follows:(1) Complex eigenvalue analysis is employed to predict the stability of vehicle disc brake system. Based on modal coupling theory, this thesis carried out the complex eigenvalue analysis on a vehicle disc brake system. The influence of the friction coefficient, the brake pressure and the wear on the stability of vehicle disc brake system is researched. The results have important guiding significances to the stability analysis and optimization of vehicle disc brake system.(2)The interval-probability uncertain model of the vehicle disc brake system is built to consider the uncertainty in the engineering and get the more reliable optimization results. According to the uncertainty caused by manufacturing error, eight parameters (the density, young modulus and thickness of back plates, the density, young modulus of friction plate and disc, brake pressure) are selected as probability uncertain parameters. Also, according to the uncertainty caused by manufacturing error and wear, three parameters (the thickness of friction plate and disc, friction coefficient) are selected as the interval-probability uncertain parameters.(3) The stability of the vehicle disc brake system with uncertainty is analyzed. To improve the efficiency of FEM analysis, sample points are selected by using Latin hypercube method in this research, and then the sampled points are subject to FEM. Next, the Kriging surrogate model of vehicle disc brake system can be built. The results show that the Kriging surrogate model was in a good fitting precision, and could be used in the analysis and optimization of vehicle disc brake system. Then this thesis conducts the stability analysis for the disc brake system with interval-probability uncertainty and the disc brake system with probability uncertainty by Monte Carlo method, based on the Kriging surrogate model. Finally, the effect of the thickness of back plates, the density and the young modulus of the disc, friction coefficient on stability is revealed.(4) The 6? robust optimization design method is proposed to analyze the vehicle disc brake system with interval-probability uncertainty. The density, young modulus and thickness of back plates are selected as design variables which are used to construct the objective function. The robust optimization result shows that the proposed approach can be used to improve the stability of disc brake system, and also improve the robustness of design result.To suppress the brake noise of vehicle disc brakes, this thesis carries out the stability analysis for disc brake system based on the uncertain model, and conducts research on robust optimization of the uncertain disc brake system. The proposed approachs have a good potential guidance and application in the engineering field of disc brake stability analysis and optimization.