Abstract: Like other sensors, ambient temperature changes significantly affect the accuracy of earth pressure cell test data. However, current test methods seldom consider the impact of temperature changes on resistance strain-type earth pressure cell readings. A study was conducted to enhance the reliability of test readings by examining temperature effects on resistance strain-type earth pressure cells. Earth pressure cells with calibration coefficients of 0−100 kN/mV, 100−200 kN/mV, and 200−300 kN/mV were placed in a temperature-controlled chamber. Using the temperature range of Jilin Province as a baseline, the study systematically examined the variations in earth pressure cell readings from −30 ℃ to 40 ℃ and proposed a temperature correction method for the readings of resistance strain-type earth pressure cells. A three-dimensional error correction method for earth pressure cell test readings was developed. The study compared three-dimensional stress values before and after temperature correction in a subgrade frost heave model test, enhancing data accuracy. The results indicate that as the ambient temperature decreases, the earth pressure cell readings increase approximately linearly; as the temperature rises, the readings decrease nonlinearly. Additionally, the hysteresis loop of the stress-temperature curve remains largely unchanged with repeated temperature cycles, though individual earth pressure cells show distinct patterns. Temperature corrections should use incremental formulas for cooling and heating processes, forming a piecewise function. This method is also applicable to three-dimensional stress testing, where temperature-induced errors in normal stress exceed those in shear stress. This study enhances the accuracy of earth pressure cell measurements and provides conditions for accurately revealing stress evolution under variable temperatures.

Key words: earth pressure cell, temperature cycling, curve fitting, temperature correction, error analysis