航天用反作用飞轮旋转变压器位置检测算法

目前，航天用反作用飞轮大多采用霍尔传感器或光电码盘进行测速。但是，霍尔传感器在反作用飞轮低速运行时精度相对较低，光电编码器的环境适应性相对较弱。基于此，提出了一种使用旋转变压器检测反作用飞轮转子位置的方法。但如果在现有飞轮控制电路中额外使用旋转变压器专用解码芯片，会导致成本大大提高，故提出了使用控制电路中的FPGA进行解码的方法。首先，介绍了旋转变压器的工作原理，通过求解反三角函数获得转子位置。其次，介绍了传统坐标旋转数字计算机(Coordinate Rotation Digital Computer，CORDIC)算法。最后，针对传统CORDIC算法无法求解完整平面角度值问题，提出了一种改进型CORDIC算法求解转子位置，并给出了一种能够减少硬件使用资源的全流水线CORDIC阵列结构。通过Modelsim仿真，证明了所提出的方法具有占用资源较少、延迟低、测量精度较高等优点，在反作用飞轮测速应用中具有良好前景。

Position Detection Algorithm of Resolver in Reaction Flywheel

At present, most aerospace reaction flywheels use Hall sensors or photoelectric encode for speed measurement. However, the accuracy of Hall sensor is relatively low when the reaction flywheel operates at low speed, and the environmental adaptability of the photoelectric encoder is weak. Therefore, a method of using a resolver to detect the position of the reaction flywheel rotor is proposed in this paper. In addition, if the special decoding chip of resolver is used in the existing flywheel control circuit, the cost will be greatly increased. Therefore, a decoding method using FPGA in the control circuit is proposed. Firstly, the working principle of the resolver is introduced in this paper. This method obtains the rotor position by solving the inverse trigonometric function. Secondly, the traditional coordinate rotation digital computer (CORDIC) algorithm is introduced. Finally, aiming at the traditional CORDIC algorithm cannot solve the problem of the full plane angle value, an improved CORDIC algorithm is proposed to solve the rotor position, and a fully-pipelined CORDIC array structure which can reduce hardware usage resources is presented. It is proved that the method proposed in this paper has the advantages of less resource consumption, low delay, high measurement accuracy by Modelsim simulation. This method has good prospects in the application of reaction flywheel speed measurement.