Time-Series Event Prediction for the Uranium Production Wells Using Machine Learning Algorithms
Bibliographic record
Abstract
ABSTRACT: The efficient estimation of the production rate in the uranium reservoir plays a vital role in the evaluation of operational performance. This paper presents the data-driven production model in the uranium field using LightGBM for the data from the Kazakhstan deposits. We focus on predicting the fault events of the well production solution. Numerical results of this investigation show that LightGBM achieves an accurate prediction with wavelet transformation. The evaluation of the model score is conducted by using metrics such as Recall and F1. With feature engineering by wavelet transformation, we obtained the recall of 0.84 and f1 of 0.89. The LightGBM model with the Morlet wavelet transformation can be useful to solve the issue of prediction maintenance of production well. 1 INTRODUCTION Uranium production has been gaining importance in recent years since it can be utilized for sustainable nuclear energy, which is also a low-carbon energy source. Due to the Covid-19 pandemic in the world, the optimization of uranium production is required to meet limited demands. In situ leaching (ISL) or in situ recovery (ISR) technology is widely used for uranium production in many countries, including the US, Canada, Russia, China, Kazakhstan. Kazakhstan has a large portion of the world’s uranium production. The main deposits utilized the ISL technology for the production places such as Inkai, Mynkuduk, Moinkum, Kanzhugan, and others (Grancea et al., 2020). ISL method is one of the key recovery technology for uranium in the Kazakhstan fields. In this method, the sulphuric acid is injected into the subsurface to extract pregnant uranium solutions to the surface (Grancea et al., 2020). Next stages, the pregnant solutions go through several chemical processes and refinement steps to recover a uranium concentration. The literature on the modeling of well production shows a variety of approaches, particularly the flow and transport simulations in the oil and gas industry. In (Regnault et al., 2015), the reactive transport model in 3D was proposed to predict the production of uranium solution. In this model, the accuracy of the result can be achieved by using detailed information on the geometrical properties. To improve computational time of the reactive transport model, parallelization with GPU was developed for the streamline-based simulation in (Tungatarova et al., 2020). Asymptotic analytical solution of the 1D flow with chemical reaction was developed by (Panfilov et al., 2016). This model was verified qualitatively by the result of laboratory experiments for the uranyl sulfate.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| 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".