Automated Analysis of DTS Data
Investigator: Dante Orta
Most fracture analysis methodologies use pressure data as the main input. However, pressure has the limitation of losing information as localized effects get smoothed out from the data. Temperature suffers from less information losses allowing for the preservation of sharp boundaries, relevant characteristics for analyzing the behavior of wells during operation and treatments Distributed Temperature Sensing (DTS) technology has increased the availability of temperature data at frequency and good spatial resolution. This fiber optic technology can be deployed in several configurations in wells and provides continuous measurements. Nonetheless, a big challenge for using DTS for analysis is the high dimensionality of the data. A well-trained person can identify the presence of fractures by observing patterns in the spatial time series plotted as ‘waterfall plots´ (Figure 1). However, this is a laborious process and does not allow for the use of temperature data in real time.The research is focused in applying Machine Learning to aid with the interpretation of DTS data. The goal is to automatically detect the presence of fractures from the data and give an estimation of flow from them. Having the capacity to automate some of the interpretation would allow for the use of fiber optic data to real time decision making.