| The rising self-driving tour has brought people a lot of fun in life,and the luggage rack has also brought a lot of convenience to travel.But at the same time its imperfect shape will increase the aerodynamic resistance and aerodynamic noise of the car,which will not only increase the fuel consumption but also reduce comfort of the car.Reducing aerodynamic resistance and aerodynamic noise has become an important part of improving automobile economy and comfort.However,the effect of changing the shape of the luggage rack on reducing aerodynamic resistance and aerodynamic noise is inconsistent.So how to weigh the relationship between the two becomes crucial.Collaborative Optimization(CO)can better constrain the two to find the optimal solution.In this paper,a common luggage rack on the market is installed on the basis of a certain SUV car model as the research object.First,resistance reduction and noise reduction are studied as two separate sub-disciplines,and the important factors affecting aerodynamic resistance and aerodynamic noise are found respectively.A model is established by using the grid deformation technology to parameterize the key parts of the luggage rack.Next step is to use CFD numerical simulation software to calculate the selected sample points,and establish a reliable Kriging response surface model.Finally,based on the approximate models of the two sub-disciplines,a collaborative optimization model is constructed to find the best low-resistance and low-noise solution.The main research contents of this paper are as follows:1.Model processing before simulation calculation.For a certain SUV model,the gap between the inside and outside of the body is filled and irrelevant details are ignored,and a luggage rack is installed.The five-layer grid encryption area is set to fully capture the flow field details,and the grid independence is verified,which also meets the required grid size for detecting the noise frequency at 5000 Hz.2.Numerical simulation calculation of the resistance and noise of the original model.11 far-field monitoring points are set up to monitor far-field noise.The resistance and noise are simulated and simulated respectively,and the flow field results are analyzed to find a total of 4 parts that have a greater impact on the resistance and noise as variables.3.Single objective optimization of resistance and noise.The optimal Latin hypercube sampling method is used to select 41 sets of sample points for 4 design variables,and 41 sets of sample points models are established using grid deformation technology,and then CFD numerical simulation is used for calculation.Then the Kriging approximate model is used to construct the response relationship between design variables and aerodynamic drag as well as between design variables and aerodynamic noise.According to the determination coefficient R2 and other methods,the accuracy of the approximate model is analyzed,and the best approximate model of aerodynamic drag and aerodynamic noise is determined.Finally,the Multi-island Genetic Algorithm is used to perform single-target optimization on the two targets to obtain low-resistance and low-noise design variables,and models are established for verification and analysis.4.The collaborative optimization of resistance and noise.Based on the approximate model established above,a collaborative optimization framework is built on Isight software to find a low-resistance and low-noise model,and the relationship between the two is weighed to find the optimal model,and simulation verification and analysis are performed.The final conclusions of this paper are as follows:(1)By single-objective optimization,resistance reduced by 1.76% with an error of0.26%.By single-objective optimization,noise reduced by 15.39% with an error of3.07%.The errors of the optimization results and the numerical simulation results were both within 5%,which has high credibility.(2)Collaborative optimization of low resistance and low noise results: aerodynamic drag reduced by 1.60% with an error of 0.17% and aerodynamic noise reduced by14.92% with an error of 3.84%.The errors of the optimization results and the numerical simulation results are within 5%,and both have a high degree of credibility.(3)In this paper,the single-objective optimization of aerodynamic resistance and aerodynamic noise and the final balance of the ratio of the two single-objectives are coordinated to obtain a lower resistance and noise model,which provides a certain reference meaning for the optimization of automobile aerodynamic shape,and can increase the reference value for selecting luggage racks with better economy and comfort. |
