From Data to Diagnosis: Exploring DTS/DAS for Early Detection of ESP Degradation

Abstract Electric Submersible Pumps (ESPs) are a important component in artificial lift systems, yet they are susceptible to various forms of mechanical degradation over time, such as bearing wear, impeller damage, and shaft misalignment. Traditionally, downhole sensors have provided limited insight into the early onset of such issues, often signaling problems only after performance is impacted. This paper explores whether Distributed Temperature Sensing (DTS) and Distributed Acoustic Sensing (DAS), typically used for flow monitoring and thermal profiling, could be leveraged as diagnostic tools to identify the onset of mechanical degradation in ESP systems. The objective is to assess the feasibility and conceptual framework of applying fiber-optic sensing as a non-intrusive, real-time surveillance method for mechanical health monitoring of ESPs. This assessment is based on DTS and DAS capabilities in the context of ESP operation. Where it is used to examine theoretical signal behaviors, potential degradation signatures, and relevant case studies from analogous applications such as flow assurance and downhole leak detection. The analysis focuses on the sensitivity of DTS to detect subtle thermal changes caused by inefficiencies, frictional heating, or flow disruptions, and the ability of DAS to pick up acoustic anomalies that may correspond to mechanical disturbances within the pump assembly. The study also considers the practical aspects of implementation, including fiber installation, data acquisition strategies, and the limitations of current interpretation workflows. This paper evaluates whether the sensing modalities have the potential to augment ESP diagnostics beyond conventional sensors. The evaluation indicates that DTS and DAS is capable to contribute to mechanical degradation detection in ESPs. Certain thermal and acoustic behaviors associated with early stage wear or misalignment may be within the resolution range of distributed sensing systems. However, signal differentiation remains a key challenge particularly in isolating mechanical signatures from fluid-related noise or dynamic flow conditions. The paper underscores the importance of advanced data analytics, and integration with surface and downhole data to extract actionable insights. This work proposes an innovative extension of distributed sensing technology into the realm of mechanical diagnostics for ESPs. It challenges the conventional paradigm of relying solely on point sensors and opens a path toward more holistic, distributed monitoring strategies. By assessing the foundational feasibility of DTS and DAS for this application, the study lays the groundwork for future research and potential integration into next-generation artificial lift surveillance programs.