针对空间引力波探测用微推力器推力测量装置在工作时受温度影响导致测量结果精度不准确的问题，提出了一种在FPGA平台实现的温度自适应Smith模糊比例-积分-微分控制器（proportional integral derivative, PID）补偿方案。该方案采用自顶向下的多级流水线设计思想，并结合MATLAB模糊工具箱进行间接实现。通过Smith预估补偿解决温度时滞问题，在Simulink中进行算法模型仿真验证后，基于Quartus Ⅱ开发平台，利用FPGA平台的内部存储器资源建立模糊PID离线查找表，间接实现模糊推理，简化了传统自适应Smith模糊PID控制器复杂的推理过程。仿真结果表明，该温度自适应控制系统功能设计正确，有效改善了传统控制器设计复杂、响应速度迟缓等问题，实现了期望的设计目标。

In order to address the problem of measurement inaccuracy in thrust measurement devices of micro-thrusters used in space gravitational wave detection caused by temperature variations during operation, a temperature-adaptive Smith fuzzy PID(proportional integral derivative) controller compensation scheme was proposed and implemented on an FPGA platform using a topdown multi-stage pipeline design approach and indirectly realised using the MATLAB fuzzy toolbox. The temperature delay problem was addressed by Smith prediction and compensation, and after verifying the algorithm model by simulation in Simulink, a fuzzy PID offline lookup table was established based on the Quartus Ⅱ development platform by using the FPGA's internal memory resources, which indirectly implements fuzzy reasoning and simplifies the complex reasoning process associated with traditional adaptive Smith fuzzy PID controllers. The simulation results demonstrate the efficacy of the proposed approach, substantiating the correctness of the functional design of the temperature adaptive control system. It is evident that the system effectively mitigates the complexity and slow response characteristics of traditional controllers, while concurrently attaining the stipulated design objectives.