Steam-Powered Sensing: Extended Design and Evaluation

Sensornets promise to extend automated monitoring and control into industrial processes. In spite of great progress made in sensornet design, deployment and operational cost impedes use of sensornet in many real-word scenarios—these challenges are so great that industries often continue with infrequent, manual observations over automation, even for key business processes. In this paper we propose two novel approaches to reduce system deployment cost, and test those approaches in a real-world monitoring of steam pipelines in oil industry. First, we make use temperature differences between the pipeline’s surface and the atmosphere reliable source of harvestable energy for sensornet operation. We demonstrate that for temperature differences of 80 ‰ or more, batteryless operation is possible using only a low-cost 40x40mm thermoelectric generation module; thereby significantly reduce costs of deployment and operation. Second, we show that non-invasive sensing can infer blockages in water and steam pipelines, and partial blockages in steam pipelines. Non-invasive sensing eliminates the need to pierce the pipeline, greatly reducing deployment cost. Finally, we evaluate our “steam-powered sensing” system in an application monitoring blockage in steam pipeline chokes in a production oilfield. To our knowledge, this is ∗This research is partially supported by CiSoft (Center for Interactive Smart Oilfield Technologies), a Center of Research Excellence and Academic Training and a joint venture between the University of Southern California and Chevron Corporation. the first industrial field sensornet deployment that employ non-solar energy harvesting. To broaden our results, we also show our approaches also apply to blockage detection in lower temperature water pipelines through laboratory experiments.