Three Approaches to Predicting ESP Pump Failures During SAGD Operations
Bibliographic record
Abstract
Abstract This paper presents a novel approach to a time scale discretization when predicting ESP pump failures at different scales. This study proves that models can be used to formalize failure predictions, prevention, and lead to optimizing the ESP's replacement and/or maintenance. The target parameters reflected two different time scale ranges. In the first approach ‘Time to Failure’ and its corresponding ‘Active Time to Failure’ were predicted. The second case excluded time periods when a well was off-line for other reasons than failure. These two targets (modeling parameters) represented low frequency events and were developed using geological or/and well geometry parameters. The Total Time to Failure model (Production Period model) based on a combined trajectory and geology data set showed acceptable and stable performance. A corresponding model with wellhead parameters summarized across each production period was introduced to complement the large scale analysis. A second group of models of higher resolution was designed to detect failures in real time. In these cases estimations for the probability of a failure at a specific time using the most recent wellhead data while excluding well's non-active time periods related to workovers and other non-productive time periods. These models used pre-processed wellhead data from a few selected wells and pads. Well data required pooling large amounts of data and developing a parameter summarization in time periods based on uninterrupted Motor Current Time Periods. These discrete time periods represented events with or without a failure depending on a reason for the current value to be zero. The probability of a pump failure was estimated using two approaches. In the first approach only the last two ‘periods’ that corresponded to non-failure and failure periods respectively were used. The second approach involved all non-failure periods leading to each corresponding failure period. The first approach overestimated the failures while the second approach overestimated the non-failure events. Initial probability models predicted events with a relatively high success rate. However, more data and additional data transformations are required to verify the practicality of our approach. More refined sub-period estimates in each Current Time Periods may help in developing improved models.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".