Precision Control Of Active Support Force Feedback System

With the continuous increase of telescope diameter, the active support technology is widely applied. Active support technology is mainly through the function of optical element back force correction due to the surface shape error caused by elastic deformation, temperature gradient and manufacturing. The precision of the active support back force is the direct factors influencing the main lens shape.This paper based on the department’s 1.2m primary mirror of thin active support project, the precision control of major research initiative to support force feedback system, by controlling the force applied to the optical components back to improve the accuracy and stability. The thin primary mirror active support for the project to take 37 floating point support way, a total of 37 road pneumatic drive units, these 37 support points are divided into six regions, each 6 to 7 points, through a DSP control board to control the force, communicate via CAN bus between the control board and the host computer.In this paper, the system carried on the correction analysis of the noise that can affect the precision of the force, and by comparing theoretical, Kalman filter to filter out the noise. According to the requirements of the Kalman filter, establish the state equation and observation equations. Kalman filter can only achieve a certain effect, but the results could not theoretically meet the system requirements. The simulation analysis was carried out on the Kalman filter, and compared with the Bessel filter and threshold filter, the final decision by median filter applied Kalman filtering algorithm of double the input signal processing was reaching, better able to filter out the noise and improve the precision force control, in theory, meet the system requirements.This paper completed the hardware circuit design of the active support force feedback system, taking into account the potential areas of application to future space, which is more demanding conditions, the system volume and quality requirements are, so this topic to TMS320F2812-based controller, to achieve control and the corresponding data processing functions; DSP internal AD for acquisition module, the sensor signal data acquisition; based on the DSP built-in CAN communication module to complete the communication function between control board and PC. While using the AD5725 DAC, mutual conversion of digital and analog signals; the design of the input signal conditioning and output signal conditioning circuit for input and output signals corresponding treatment; with the corresponding power supply circuit to ensure that the control panel normal work. This article has also carried on the correlation analysis of the sensor accuracy, input signal processing precision, DA conversion accuracy and electrical converter accuracy, and the combination of the theory has carried on the related calculation precision.On the basis of the completion of the hardware system, this paper is based on PI control algorithm, programming the DSP board control procedures, including the function of data acquisition, control, AD sampling correction, input correction, electrical converter drive and communication. At the same time, based on Matlab GUI control, carried out the human-machine interface design.