Development And Identification For An Airborne Radar Servo System

This thesis studies and develops the electronic and electrical parts of an airborne radarservo system and its key parts’ homebred change. The main content includes:1, The system’soverall design based on the servo system principle;2, The development of PWM drive;3, Thesystem hardware and software design;4, Upper PC communication module;5, Identificationof system model to provide basis for setting controller parameter and to accelerate thisprocess.The electrical part of Servo system consists of controller module, drive module,detection module and communication module. Controller module with single-chipmicrocomputer as the control core, based on the PID control algorithm and three-loop controlprinciple, mainly relies on the feedback design of position loop, speed loop and current loop.Drive module is composed of PWM signal generator and H bridge chip, and PWM generatoris realized by FPGA. Detection module is formed by photoelectric coder, FPGA and currentdetection simulation circuit. While the current is detected by tandeming sampling resistor,then through A/D transmitted to the control module. Communication module connect servosystem with PC by the serial port.The work of system identification includes:selection of the right identification inputsignal,calculation of the proper sampling interval and acquisition of object output data. Mseries as identification input signal is produced by the FPGA;Acquisition of the encodersignal is considered as the output identification signal. In upper computer, the collectedidentification input and output signal were used to make model estimation and analysis byMATLAB toolbox. Identification methods including correlation method and least squaremethod were used to achieve non parameter model the impulse response, and thencorrelation to least square method is used to calculate the parameters of the model thesystem transfer function. By AIC criterion the structure identification is taken to derive thesystem order. The comparison between the identified model output curve and the actualacquisition curve verifies the validity of the model.Based on the above work, the debugging of the closed loop system was done. The resultsshow that system function is normal with communication running well.