Design and Application of High-Precision Temperature Measuring Instrument for Ice Cover Profile of River Based on the Resistance Residual Compensation Method

Xiao Deng, Yuanguo Cheng, Lingao Li, Jian Feng, Liqin Cui, Chao Du, Li Zhang, Jianmin QinIEEE Access 2020 · DOI

Monitoring the vertical temperature profile of river ice is essential for water conservancy safety and understanding thermodynamic shifts in high-latitude regions. These "temperature chains" identify boundaries between air, ice, and water based on thermal conductivity. However, the Negative Temperature Coefficient (NTC) thermistors typically used are highly non-linear, and their accuracy often degrades significantly at the extreme low temperatures (-50°C to 20°C) required for polar research.

Methodology and System Design

The researchers developed a high-precision instrument utilizing 100 PS303J2 NTC thermistors, controlled by an MSP430 low-power microcontroller and a high-accuracy LTC2983 temperature-measuring chip.

Schematic of the temperature measurement circuit featuring NTC thermistors and the LTC2983 chip.

The core innovation of this work is the Resistance Residual Compensation (RRC) method. Unlike standard calibration techniques that attempt to correct temperature output directly, RRC addresses errors in the resistance domain. The process involves:

  1. Initial Fitting: Using a standard empirical formula to relate resistance ($R$) to temperature ($T$).
  2. Residual Analysis: Calculating the "resistance residual" ($r$)—the difference between measured resistance and the formula's prediction.
  3. Polynomial Correction: Applying a second-order polynomial ($r = a + bR + cR^2$) to correct this systematic error.
  4. Final Model: Integrating this correction back into the logarithmic temperature equation to create a self-correcting model.

Experimental Validation and Field Results

The instrument was validated through laboratory calibration using a high-precision low-temperature thermostat and field observations at the China-Russia border on the Heilongjiang River.

Temperature residuals across the -50°C to 20°C range after applying the RRC method.

The RRC method achieved a temperature accuracy of 0.1 K. When compared against seven established mathematical models (including Steinhart-Hart and various Hoge equations), the RRC method demonstrated the lowest average fitting residual ($\Delta T_{avg} = 0.039 K$).

In field applications, the device successfully tracked the Heilongjiang River's ice growth from 40 cm in mid-December to a peak of 100 cm in March. It accurately captured the transition through growing, stabilizing, and melting phases, even identifying complex phenomena like the "cold intermediate layer," where a pocket of cold remains trapped deep within the ice as the surface warms during the day.

Conclusion

The RRC method significantly enhances the utility of affordable NTC thermistors for high-precision cryospheric research. While currently optimized for the PS303J2 sensor, future work will focus on generalizing this polynomial correction to other thermistor models and conducting comparative field tests with alternative technologies like Fiber Bragg Grating (FBG) sensors.

Summary updated

BibTeX
@article{deng2020design,
  author = {Xiao Deng and Yuanguo Cheng and Lingao Li and Jian Feng and Liqin Cui and Chao Du and Li Zhang and Lin Zhang and Jianmin Qin},
  title = {{Design and Application of High-Precision Temperature Measuring Instrument for Ice Cover Profile of River Based on the Resistance Residual Compensation Method}},
  journal = {IEEE Access},
  year = {2020},
  doi = {10.1109/access.2020.2964144},
}
Figure