Real-Time Predictive Temperature Measurement in Oil Pipeline: Modeling and Implementation on Embedded Wireless Sensing Devices

Abstract

In pipeline management, the advent of Internet of Things (IoT) technology enables effective pipeline maintenance with the deployment of embedded sensing units, the basic requirement of which is to make accurate and real-time measurement. Practically, it is impossible to directly measure the internal temperature of the pipe by laying conventional sensors in the oil. This work proposed a novel indirect testing framework to perform real-time and accurate temperature monitoring in oil pipes by solving the inverse problem based on pipe wall's temperature field. The contribution includes three aspects: 1) A rigorous theory was developed to validate the principle of equivalence for heat transfer in pipes. A single-layer boundary-equivalent diffusion model with effective thermal parameters exhibited excellent performance in predicting temperature change in a multilayer structure by simulation. 2) We established a microprocessor-enabled predictive temperature measurement modality to precisely determine the time-dependent temperature evolution in a multilayer composite pipe based on explicit analytical expressions. 3) The correctness and applicability of the modality were verified experimentally on a real oil pipe and a commercial device. The results show that precision of the predictive measurement modality reaches 0.3 K and 0.002 K/s for stable and linear-rising internal fluid temperature, respectively.