ESP "healthy" operational conditions characterization

• Grant number: 24/00056-0
• Support Opportunities: Scholarships in Brazil - Master
• Start date: March 01, 2024
• End date: February 28, 2026
• Field of knowledge: Engineering - Mechanical Engineering - Mechanics of Solids
• Agreement: Equinor (former Statoil)
• Principal Investigator: Alberto Luiz Serpa
• Grantee: Paulo Yoshio Kuga
• Host Institution: Faculdade de Engenharia Mecânica (FEM). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Company: Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Mecânica (FEM)
• Associated research grant: 17/15736-3 - Engineering Research Centre in Reservoir and Production Management, AP.PCPE

Abstract

This project aims to characterize the conditions considered normal for operation, obtaining patterns that represent the electrical submersible pump (ESP) operation without anomalies, that is, the case considered "healthy."Even in a normal situation, operating data may present recording problems such as time segments in which data was not stored, loss of information from specific sensors for a particular time, significant measurement uncertainties, and other aspects. These situations must be mapped to separate possible problems and data characteristics from conditions of changes in the behavior of the pumping system. It will seek to identify and characterize the cases considered regular operation, without failures or damages, and verify the usually measured signals, identifying the characteristics that can be regarded as false alarms for failures or damages.Operating data such as temperature, pressure, current and vibration in the pump, and temperature and pressure at the wellhead, would be analyzed in the time domain in terms of their main characteristics regarding continuity, lost records, and corrupted segments, looking for to identify aspects of the data that may indicate changes in operating conditions, but which effectively represent a condition still considered normal for operation.Frequency domain analyses would also be carried out to characterize typical spectra and spectrograms of the usual and considered healthy operating conditions, as well as the effect of data recording problems on these responses.The signals' processing must consider the difficulties associated with larger datasets, which can present significant measurement variability.Given the complexity and variability in the process, this project is important in identifying graphic and visual patterns, which allow saying if the operation is considered normal, or on the contrary, which alerts to the possibility of false alarms due to the characteristics of the measured data, or situations of intrinsic process variability.