Research On Modeling And Control Methods For Large Transonic Wind Tunnel

2.4m Intermittent Transonic Wind Tunnel is the only ground simulation platform that can test transonic aerodynamic characteristics of large-scale aircraft models in China.It plays an irreplaceable role in developing new aircrafts(such as C919,Yun20 and so on).2.4m wind tunnel simulates flying environment by generating high-speed flow field,so accuracy of flow field directly affects development level and technical performance of aircrafts to be developed.The wind tunnel has a complex structure,involves three-dimensional gas flow,and has the characteristics of non-linearity,multi-variable coupling and so on.These characteristics make it difficult to establish an accurate model for the wind tunnel.Due to the limitation of theory,algorithms and experiment conditions,the wind tunnel control system still depends on the model-free PID control strategy.This leads to low control accuracy of flow field in the test section and long regulating time,which is not allowed for the test requirements of developing new aircrafts.From the point of view of practical requirement,it is necessary to study specific modeling and advanced control methods for the transonic wind tunnel system and then to achieve high-precision and rapid control of wind tunnel flow field.This is significant for improving test capability of large-scale transonic wind tunnels and reducing energy consumption during wind tunnel tests.This dissertation aims to achieve high-precision modeling and control for transonic wind tunnel flow field.The operation mechanism and test characteristics of the transonic wind tunnel are analyzed in detail.Then,on the basis of considering test cost during tuning control system and safety of wind tunnel equipment,this dissertation studies modeling,simulation and advanced control strategy for the wind tunnel system.The main research work in this dissertation is summarized as follows:(1)Designing a new wind tunnel flow control structure.The new control structure takes the main exhaust valve and the grid finger as the control variables and chooses the total pressure and the static pressure as the controlled variables.The Mach number is indirectly controlled by regulating the total pressure and the static pressure.This control structure reduces the nonlinearity between the Mach number and the total pressure and static pressure,and is conducive to achieve accurate control of the flow field in the test section.(2)Studying advanced modeling and identification methods for the wind tunnel system,and building a wind tunnel system simulation experiment platform.1)Considering the problem that the wind tunnel has no running data in new working conditions,the step response data model parameter identification method,which is based on area method,is proposed.The optimal order of each part in the transfer function model is determined based on the minimum RMSE criterion.The resulting model can be used to simulate the flow field characteristics in new wind tunnel test conditions and thus reduce tuning cost of the new wind tunnel control system;2)A block-oriented model,which incorporates the nonlinear characteristics of the actuators in the wind tunnel system,is established for the typical wind tunnel working conditions.The main exhaust valve and the choke finger are considered as independent blocks,and Hammerstein model(H model)with dead-zone input is used to describe the dynamic characteristics of two parts.Then,the wind tunnel flow field is modeled as a two-input-two-output(TITO)coupled dynamic linear model.The entire wind tunnel system is finally modeled as a nonlinear multivariable system consists of the two H sub-models followed by a dynamic linear model.The ARLS method is used to identify the model parameters of the nonlinear block.The FFRLS method is used to identify the parameters of the linear wind tunnel flow field model.Practical tests shows that the proposed model has higher prediction precision and good real-time performance,and can accurately reflect the dynamic characteristics of the wind tunnel system.The block structure model can simulate the wind tunnel under typical working conditions with high accuracy.3)Building a wind tunnel system simulation experiment platform.The high-precision simulation wind tunnel system is constructed by using the constructed mode,which provides a high-precision test platform for controller test and verification.The control strategy of wind tunnel flow field is studied based on the platform.The designed controller is firstly tuned and optimized in the simulated wind tunnel and then tested in the actual wind tunnel,which reduces tuning cost of the controller and guarantees safety of the wind tunnel equipment.(3)For different wind tunnel test conditions,proposing a non-minimum state space(NMSS)model-based preview control strategy and the multi-model preview control strategy.1)To overcome the difficulty of online calculation of the non-linear block model based controller,a control-oriented NMSS model for steady flow field is proposed.The NMSS model facilitates engineering application of advanced control algorithms,e.g.optimal control.The stable weighted recursive least squares(Stable_WRLS)algorithm is used to identify the model parameters,which can guarantee model stability.All the state variables in the NMSS model are measurable historical data at current and past instants,which avoids design of complex state observers.The NMSS model facilitates the design of wind tunnel flow controller by advanced control algorithms.2)For varying angle of attack tests,a flow field predictive controller is designed for a single condition in the wind tunnel,which can effectively reject the flow field disturbance resulting from varying angles of attack and realize high-precision and fast control of the flow field;3)For characteristic of flow field changing significantly in the varying Mach number test,a multi-model preview controller for multi-conditions is designed.It can effectively restrain the influence of system parameters changing significantly and setting value varying during test processes,and can realize high-precision control of flow field and non-disturbance switching.(4)On the basis of theoretical research,the real-time preview control system of wind tunnel flow field has been developed.On the basis of the original wind tunnel control system,the PXIe-8135 used as the hardware platform of the flow field controller is combined with NI Lab VIEW software,and the optimal preview control algorithm and the multi-model optimal preview control algorithm are implemented by LabVIEW software,which achieves the engineering application in the actual wind tunnel flow field control system.The validity of the real-time preview control strategy for the wind tunnel flow field is verified by the actual tests in the wind tunnel.