Minimizing temperature drift errors of conditioning circuits using artificial neural networks

Temperature drift errors are a problem that affect the accuracy of measurement systems. When small amplitude signals from transducers are considered and environmental conditions of conditioning circuits exhibit a large temperature range, the temperature drift errors have a real impact in systems accuracy. In this paper, a solution to overcome the problem of temperature drift errors of conditioning circuits is proposed. As an example, a thermocouple-based temperature measurement system is considered, and the stability of its conditioning circuit (AD595) is analyzed in two cases: with and without temperature drift error compensation. An Artificial Neural Network (ANN) is used for data optimization and a Virtual Instrument, using GPIB instrumentation, is used to collect experimental data. Final results show a significant improvement in the accuracy of the system when the proposed temperature drift error compensation technique is applied to compensate errors caused by temperature variations.