基于低频滤波法的T-TDR含水率测量方法研究

为了解决T-TDR(Thermo-time domain reflectometry)探针短,土壤含水率测量效果差等技术问题,提出了一种低频滤波法LFF(Low frequency filter)进行TDR波形分析的计算模型。通过对TDR波形快速傅里叶变换,得到组成TDR波形的频谱,对其低频滤波后重新快速傅里叶反变换,可实现LFF变换。试验通过4种土壤,每种5个含水率的20个土样进行测试,通过LFF法和传统WINTDR法对比,结果表明LFF法具有较好的计算结果,其与干燥法的决定系数R2达0.983 5。LFF法不仅有效提高了T-TDR短针土壤含水率的测试精度,还简化了土壤含水率测试校验程序,提高了测试效率。

Measurement of Soil Moisture with T-TDR Probe Based on LFF Method

In order to improve the accuracy of soil moisture measurement by T-TDR (Thermo-time domain reflectometry) probe because the needles of T-TDR were too short, this paper presented a new TDR waveform analysis method named LFF (Low frequency filter). The TDR waveform was transformed by fast Fourier transform algorithm, and then the spectrum of TDR wave was obtained. The LFF transformation could be realized by filtering the low frequency of TDR wave spectrum and transforming it by inverse fast Fourier transform. Four kinds of soil samples prepared by five different volumetric water content with twenty specimen, were used in the experiment. Through comparing the LFF method and the traditional WINTDR method with the oven drying method respectively, LFF method showed better results. The correlation coefficient of LFF with the oven drying method was R2=0.9835. LFF method could improve the soil moisture measurement accuracy of T-TDR and simplify the test validation procedures and improve the test efficiency.