Aerodynamic Surrogate Model Design And Control Allocation Of Flying Wing Unmanned Aerial Vehicle For High Efficiency

Flying wing unmanned aerial vehicle(UAV)is a kind of tailless aircraft with Blended-Wing-Body configuration,which can reduce fuel consumption,improve en-durance,aerodynamic and stealth performance.Therefore,the flying wing is one of the important development directions of aircraft.However,the multi-control surface con-figuration of the flying wing makes the flight control system become over-drive system,which brings the problem of control redundancy.An effective method is to design the flight control system based on the cascade architecture of "flight control law+control allocation",and solve the problem of control surface redundancy by control allocation method.In the design process of control allocation,factors such as allocation precision and control energy consumption should be considered to achieve efficient allocation results.In this paper,a high-efficiency control allocation method for multi-task phase with high allocation accuracy,low control energy consumption for flying wing UAV is stud-ied.The nonlinear aerodynamic characteristics,multi-axis control coupling and interac-tions of the control surface of the flying wing UAV were analyzed,and the key problems in the design of Kriging and Co-Kriging aerodynamic surrogate models were studied.Taking control energy consumption as a significant factor,the nonlinear control allo-cation method,multi-objective control allocation method and a reconfigurable control allocation method based on the aerodynamic surrogate model were proposed,and the proposed allocation method was applied to a flying wing UAV with morphing control surfaces.This dissertation carries out several aspects of research work as follows:(1)In order to solve the problem of low energy consumption and high precision nonlinear control allocation for flying wing UAV,the nonlinear aerodynamic control characteristics,multi-axis control coupling characteristics and interactions of elevons and split drag rudder were analyzed.The nonlinear approximate accuracy of several typical surrogate models was comprehensively analyzed and compared,and the high precision Kriging model was used to establish the aerodynamic surrogate model of the control surface of flying wing UAV.Based on the aerodynamic surrogate model,the nonlinear control allocation problem of the flying wing UAV was transformed into a constraint optimization problem with the goal of allocation accuracy and control energy consumption,and an efficient nonlinear control allocation method based on the Kriging aerodynamic surrogate model was proposed.The allocation performance of genetic algorithm and gradient accelerated particle swarm optimization algorithm is analyzed,and a hybrid optimization algorithm is proposed to solve the control assignment problem based on the two algorithms.(2)On the basis of reducing control energy consumption,considering that flying wing UAVs will go through multiple flight stages,an efficient multi-objective control al-location problem based on the multi-fidelity aerodynamic surrogate model is studied.In order to improve the accuracy of aerodynamic modeling and reduce the calculation cost,Co-Kriging multi-fidelity surrogate model was used to model the control surface of fly-ing wing UAV.An optimal Latin hypercube sampling method with non-uniform interval partition is designed to obtain the initial sample points of the high confidence surrogate model.A parallel point strategy based on root mean square error and prediction gradi-ent and a multi-termination criterion for global modeling accuracy are proposed.The requirements of control energy consumption,allocation precision,lift coefficient and drag coefficient of flying wing UAV in different flight stages were analyzed.A math-ematical model of multi-objective control allocation was established,and the synthetic function method and layered optimization multi-objective particle swarm method were proposed to solve the problem of multi-objective control allocation.(3)Considering that complex multi-control surface configuration will increase the probability of control surface system failure,the reconfigurable control system of flying wing UAV under typical control surface failure is designed based on Kriging control allocation architecture.The failure modes of the control surface,actuator and sensor in the control system are analyzed,and the four typical failure modes of the control surface,such as damage,jamming,loosening and floating,and median offset,are mainly studied.By analyzing the influence of the control surface faults on the handling performance,the failure mathematical model under the corresponding fault state is established.Accord-ing to the different control surface fault forms,the original Kriging model was modified in corresponding fault states,and the reconfigurable control system of flying wing UAV was designed based on the modified Kriging aerodynamic surrogate model.(4)The application of morphing control surface technology to flying wing UAV can further improve the aerodynamic performance of control surface and reduce control energy consumption.The feasibility of using the deformed trailing edge as a morphing control surface was analyzed,and a deformed wing trailing edge structure was designed based on the flexible skinning technology.The Attainable Moment Sets(AMS)between the morphing control surface and the conventional control surface of the flat wing is compared and analyzed to verify the effect of improving the aerodynamic performance of the morphing control surface.Then,the morphing control surface was applied to the flying wing UAV,and the aerodynamic characteristics were analyzed by using the 3D panel method.The Kriging aerodynamic surrogate model of the morphing control sur-face was established,and the control allocation of the flying wing UAV with morphing control surface was preliminarily designed based on the above allocation architecture.The validity of the proposed control allocation method is verified by the performance test of open-loop allocation.To sum up,this paper takes the allocation accuracy and control energy consump-tion as the basic objectives,proposes the control allocation method based on the aero-dynamic surrogate model,and designs the Kriging and Co-Kriging models for global approximate accuracy.The nonlinear control allocation problem,multi-objective con-trol allocation problem,reconfigurable control problem and morphing control surface control allocation problem of flying wing UAV are studied.The research work in this paper can provide guidance for the design of the flight control system of the flying wing UAV,and also provide solutions for the control problems of other over-drive systems.