Aerodynamic configuration optimization method based on flow field prediction

Abstract

The invention discloses an aerodynamic configuration optimization method based on flow field prediction. The method includes subjecting flow field pressure distribution directly to approximate modeling; providing a flow field approximation model deeply coupled with the aerodynamic optimization problem of the aircraft; realizing efficient utilization of a large amount of flow field data generated by the high-precision flow field simulation model, thus reducing the number of simulation calling times of the high-precision model of the sequence approximate optimization method, improving the quality and efficiency of aircraft aerodynamic configuration optimization, and greatly improving the optimization efficiency of engineering optimization problems. And compared with a conventional sequence approximate optimization method and a stochastic evolution algorithm, the effectiveness and the high efficiency of the method are verified, and the aerodynamic configuration optimization method provided by the invention has important application value for aircraft aerodynamic configuration fine design.

Description

technical field [0001] The invention relates to the technical field of aircraft aerodynamic shape optimization design, in particular to an aerodynamic shape optimization method based on flow field prediction. Background technique [0002] Aerodynamic shape design is an important part of aircraft design and has an important impact on overall performance. With the improvement of aircraft performance requirements, the design of aircraft becomes more complex, which puts forward higher requirements for the fine design of aerodynamic layout. The traditional aerodynamic engineering estimation model is difficult to apply in many occasions, and high-precision numerical simulation is imperative; the traditional evolutionary algorithm usually requires thousands of iterations, which is unacceptable for the time-consuming high-precision simulation model, and the approximate optimization method came into being , in order to achieve a balance between aerodynamic design accuracy and effici...

AI Technical Summary

Problems solved by technology

This method has the following disadvantages: 1) The determination of the target flow field distribution requires the designer to have a deep understanding of the design target and rich design experience, and the design quality depends heavily on the selection of the target aerodynamic characteristics; 2) It is difficult to deal with the aerodynamic , geometric constraints, and performance constraints at non-design points

This method has the following defects: 1) The biggest limitation of the gradient-based optimization method is that it is easy to fall into the local optimum and it is difficult to find the global optimum. To a certain extent, it limits its scope of application and further development; 2) Due to its random search characteristics, heuristic algorithms generally have the characteristics of slow convergence speed, and the optimization process requires thousands of iterations of the calculation model. For engineering optimization problems requiring high precision, combining time-consuming simulation models will make the amount of calculation unacceptable, which is also the biggest obstacle to the application of heuristic algorithms in engineering, which greatly limits the scope of application; 3) Based on proxy models (also known as approximate Model Surrogate Model, or Metamodel) optimization method, using an approximate model with high computational efficiency instead of a high-precision simulation model, effectively improving the efficiency of optimization design

In this process, the simulation model is regarded as a "black box", and the approximate model only predicts the input-output relationship of the "black box", which cannot make full use of the large amount of data information generated by the time-consuming simulation model, which seriously weakens the ability of the proxy model to capture the real The ability of the essential characteristics of the model greatly limits the further improvement of the efficiency of the approximate optimization method; for specific engineering optimization problems, such as the aerodynamic design optimization problem, it is a further step to fully utilize the simulation data through the strong coupling between the approximate optimization method and the calculation model. An Efficient Way to Tap the Efficiency Potential of Approximate Optimization Methods

Images