Research On Active Adaptive Noise Reduction Of Turboprop Driven Aircraft Based On Synchrophasing Control

The noise level caused by the propeller operation is one of the main objectives of the turboprop aircraft airworthiness certification,and the high noise level will also damage the hearing level of the pilot and passengers.Therefore,the noise reduction of turboprop aircraft is particularly important.Active noise reduction control technology can reduce noise level more effectively than passive noise reduction.In various active noise controls,the multi-propeller phase synchrophasing control is highly feasible,and the noise reduction effect is remarkable,and the aircraft weight is not additionally increased.It is widely used for propeller aircraft noise reduction.However,there are still some problems unsolved with the development of propeller synchronphasing control.This article focuses on the topic of ensuring and improving the noise reduction effect in complex environments.Based on the noise reduction mechanism,model identification,signal processing,and experimental verification,the following research contents are specifically carried out:Firstly,the process of establishing the noise source model is analyzed and simplified.The current propeller noise modeling process does not consider the difference of the speed,dynamic characteristics and phase of multiple propellers.The joint computational fluid dynamics and the modified harmonic method is proposed to establish the multi-propeller noise model.The computational fluid dynamics technique is used to calculate the multi-propeller performance,and the mechanical properties of the propeller under different working conditions are calculated.On this basis,the time parameters in the harmonic method modeling method are modified,the multi-propeller phase delay time is added,and the mechanical properties of the multi-propeller are used to calculate the sound pressure level.This scheme makes it easy to establish a multi-propeller noise source model,and fully considers the different mechanical properties and phase differences between the multiple propellers.The effects of flight conditions and number of propellers on the noise model are studied.Under different heights,Mach number,engine speed and different number of propellers,the noise distribution when the phase angle is changed is calculated,and the noise generation and propagation mechanism in the case of no chamber is discussed.Then,aiming at the problem of limited measurement position of microphone in the actual measurement environment,a method of widening the noise measurement range by using virtual sensing technology is proposed,including designing and constructing a propeller synchrophasing control noise reduction ground experiment platform to study propeller synchrophasing.In order to avoid the drawback the traditional master-slave control architecture,such as the delay of control signal communication and the weak anti-interference ability,all-slave control architecture is proposed to replace the master-slave control architecture.The simulation and comparative study of the two different architectures are carried out,and then all-slave control architecture is deployed to the propeller synchrophasing control experimental platform.The factors affecting the accuracy of the propeller signature model are studied,and a criterion for evaluating the quality of the measured noise signal is proposed.In this paper,a minimum fluctuation extraction algorithm of acoustic signal is proposed.Wavelet filtering and three parameter sine fitting algorithm are used to filter and reconstruct the collected microphone noise data.The minimum fluctuation segment is selected as the input of propeller signature model.The minimum fluctuation data obtained by this method can improve the quality of acoustic signal to the greatest extent,so as to improve the accuracy of identification of the propeller signature model.Then the forward differential prediction technique and the stochastic optimal diffusion sound field prediction technology are deployed into the actual measurement environment,which expands the measurement range of noise and provides the possibility of noise measurement at key measurement points in the real cabin.The ground propeller synchrophasing control experiment with the cylindrical scaled fuselage is carried out,including the necessity and feasibility of introducing the small cylindrical scaled fuselage into the ground experimental platform.The propeller synchrophasing control experiment platform with the fuselage is designed to verify the noise reduction effect in the real cabin of the propeller synchrophasing control.In the presence of possible frequency leakage and fence effect of the noise signal spectrum,a mixed convolution window is proposed,and the harmonic parameter correction calculation is optimized based on nested cubic spline function.FFT harmonic analysis based on bimodal spectral lines interpolation is adopted in this thesis to avoid the error caused by the traditional FFT analysis which can’t satisfy the integer period in signal sampling.Thus,the accuracy of traditional FFT for signal analysis is improved and the frequency spectrum leakage and interference of noise signal are restrained.Therefore,the identification accuracy of propeller signature model is further improved.On this basis,the virtual microphone measurement experiment in the cabin is carried out,which further verified the feasibility of the virtual sensing technology.For the problem that the pitch angle control loop is susceptible to external disturbance and the control precision is difficult to guarantee,a scheme of using the ADRC instead of the traditional PID technology to control the propeller pitch angle is proposed.The influence of different parameters on the performance of the controller in the three parts of the ADRC is studied.The conclusion that the extended state observer parameters are the core of the entire controller parameter tuning is given.A method of calculating the gain of the observer using the non-dominated sorting genetic algorithm is proposed,which reduces the influence of parameter redundancy and randomness on the system.On this basis,the parameterized ADRC is deployed into the propeller engine component model,and numerical simulation is carried out to verify that its anti-disturbance effect is better than traditional PID technology.A new phase angle control command logic is proposed,which optimizes the phase angle control path.The simulation shows that the new logic makes the control time shorter and phase route smaller.Finally,the accomplishment on this thesis is summarized,and the future development focus and possible challenge of the further research are prospected.