分布式光纤喇曼测温系统采用了线性逼近的方法进行温度解调,并通过灵敏度分析法获取了相应的线性解调参数。为了提高系统的精度,在信号累加平均去噪的基础上,采用了卡尔曼滤波的方法,通过对前一时刻的估计值与当前测量值的递推,实现了对存在测量噪声的温度信号的最优化估计,完成了对现场采集信号的实时更新和处理。实验验证了该滤波方案的可行性,随着外界环境温度的变化,测温系统能够得到合理的温度曲线,测量误差小于1℃。系统响应性良好,性能稳定,能够适应复杂的环境变化。 The distributed fiber Raman temperature measurement system uses a linear approach to temperature demodulation, and obtains the corresponding linear demodulation parameters through sensitivity analysis. In order to improve the accuracy of the system, a Kalman filter method is used on the basis of cumulative average signal denoising. By recursive estimation of the previous time and the current measured value, the temperature signal Optimized estimates, completed the acquisition of real-time signal on-site updates and processing. The feasibility of this filtering scheme is verified by experiments. With the change of ambient temperature, the temperature measurement system can get a reasonable temperature curve, the measurement error is less than 1 ℃. System responsiveness, stable performance, able to adapt to complex environmental changes.