Precision Tracking System Nonlinear Modeling Control And Verification System Development

With the development of science and technology,the precision requirements of various engineering fields are getting higher and higher,and the performance of precision tracking systems has become one of the focuses of research.In this paper,the fast reflection mirror(FSM)is used as the research carrier.Firstly,the white box model with non-linear links of the controlled object is established by the method of theoretical derivation.For the differences between the theoretical model and the actual object,a plurality of experimental schemes are designed to obtain the measured frequency characteristic curve of the controlled object and the response curve under different signals as the basis for model correction.Using the method of system identification,the linear part of the theoretical model is corrected based on the frequency characteristic curve,and the actual measured value of the nonlinear parameter is extracted based on the response curve,and the theoretical value is mutually verified.At the same time,the same noise model was obtained using two methods based on experimental data.Finally,an essentially non-linear white box model with controlled object noise is established.According to the measured frequency characteristic curve of the controlled object,a black box model of transfer function composed of typical links was obtained by using the method of artificial fitting.Both models have their own advantages in control design and can be flexibly selected.Adopting advanced control,lag correction,notch control and other control strategies and their combinations,the closed-loop control system is designed.For the ubiquitous resonance problem in FSM,two control schemes are used.The resulting closed-loop system bandwidth can meet the design expectations of 500 Hz,even far beyond 500 Hz.The analysis compares the performance difference between the linear system and the nonlinear system under the same controller.It is concluded that the control limit is greater than the physical limit,and the nonlinearity is the key factor that restricts the system performance.Specific analysis of the physical limit of the system performance constraints.According to the size of the input signal,the linear and nonlinear regions of the system work area are divided,and the reason why the bandwidth of the system under the small signal is wider is explained.The influence of saturation nonlinearity on the transient performance of the system is analyzed.Comparing the steady-state accuracy of the systems under different controllers,the current system limit accuracy is 1.6?,and the sensor measurement noise is a factor that limits the steady-state accuracy.For this measurement noise,the control strategy cannot be suppressed and measures need to be taken from the root of the noise.The maximum tracking frequency that the system can achieve under the input of different amplitude sinusoidal signals is analyzed,the contradiction between the tracking performance and the physical limit existing in the nonlinear system is revealed,and the conclusion that the control limit is greater than the physical limit is verified.The joint control simulation was performed on the two axes and the real-time and radial errors of the circular trajectory tracked by different control systems were obtained.In order to verify the analysis method and conclusion of this paper,a single-degree-offreedom turntable was used as the load,and a three-closed-loop servo verification system driven by a DC motor was developed.The design of the hardware and software system was completed,and the function of the servo system was realized.A detailed analysis and solution was made to the key issues in the development process.Drivers were designed using two schemes,and the performance differences between the two were compared.The current sampling conditioning circuit was designed and implemented by Multisim simulation and verified by actual tests.The correctness of the circuit design is analyzed;the effects of the PWM wave frequency and resolution on the DC motor control performance are analyzed;the performance differences of the two encoder pulse signal processing schemes are compared,and the STM32 orthogonal decoding mode is finally used to effectively filter the measurement noise.Improve the measurement accuracy;complete the C language implementation with three closed-loop PID algorithm with integral separation and integral anti-saturation.Finally,the system is tested and a set of actual system tracking response curves are given to verify the realization of the servo system function.