Sensorless Vector Control Of Electric Tail Rotor Of Unmanned Helicopter Technical Study

Due to its advantages of low cost,flexible use and no site restrictions,small unmanned helicopter has been widely used in many fields,such as fire rescue,emergency medical treatment,combat communication,exploration of minerals,agricultural production and so on.It has become an essential product for both military and civil use.The traditional mechanical driven tail rotor has complex structure and low reliability.Due to its coupling with the main rotor,the speed of the tail rotor and the main rotor can only be a fixed multiple relationship,which reduces the flexibility of the helicopter.Using electric tail rotor instead of mechanical drive tail rotor can solve the above problems.In order to improve the control performance of electric tail rotor,a control system of tail rotor motor with fast response,low torque ripple,good stability and robustness is designed in this thesis.Firstly,the BLDCM with good speed regulation performance and high power density is selected as the driving motor.However,in order to reduce cogging torque in the manufacturing process of BLDCM,the methods such as inclined slot,inclined pole,fractional slot and reasonable design of pole geometry are adopted,which make the back EMF of BLDCM closer to sine wave.For the BLDCM whose back EMF waveform is quasi sine wave,using sine wave drive can effectively reduce the torque ripple and running noise.In order to obtain accurate rotor position information,a complex coefficient filter is proposed to replace the low-pass sliding mode observer to filter the back EMF signal.Because the complex coefficient filter has no amplitude attenuation and phase lag at the center frequency,through parameter calculation,the appropriate center frequency coefficient is set,so that the frequency of the back EMF is equal to the center frequency,achieving better filtering effect.In order to further reduce the chattering of the system,the saturation function is used instead of the symbol function to make the control continuous and reduce the jump of the sliding mode controller near the switching surface.By comparing the back EMF signal and speed waveform of the improved and the improved sliding mode observer,it is found that the improved sliding mode observer has higher detection accuracy.Furthermore,in order to improve the dynamic response speed and anti-interference ability of the control system,a control strategy of adding RBF neural network to the speed outer loop of vector control is proposed.The RBF neural network has the advantages of fast learning speed and strong non-linear expression ability.According to the gradient descent method,the parameters of PID controller are adjusted in real time to get better control parameters,which can reduce the response time and improve the anti-interference ability of the system.Then the simulation results show that the response time and antiinterference ability of the system with RBF neural network are significantly improved.Finally,the above three aspects of the vector control of BLDCM are integrated,and the simulation results show that the improved vector control has smaller torque ripple,faster dynamic response and better robustness.