Evaluation and Prediction of Hydraulic Fractured Well Performance in Montney Formations Using a Data-Driven Approach
Bibliographic record
Abstract
Abstract The Montney tight formation, located in Western Canadian Sedimentary Basin (WCSB), is becoming an important component of hydrocarbon sources in Canada. Horizontal drilling and multistage stimulation techniques have been successfully applied to exploit Montney formations. Selecting appropriate completion and stimulation designs are crucial to maximize the well productivity and/or oil recovery in Montney. However, it is still challenging to predict after stimulation productivity of a target well based on different stimulation parameters. Data-driven approaches, such as neural network, can be applied to evaluate hidden correlations between stimulation designs and well productions. In this study, a comprehensive data mining technique, which integrates the cluster analysis, kernel PCA and DE-based ANN technique, is successfully developed to evaluate and predict the hydraulic fractured well performance in Montney formation. Fracturing operational data and well after stimulation productions of 1521 horizontal wells in Montney tight formations are first collected and classified into different groups using three clustering algorithms. Data-driven neural network models optimized by differential evolution (DE) algorithm are then trained using the stimulation parameters and cumulative productions. More specifically, fracturing operational parameters collected include operational time per stage, well completion strategy, types of fracturing base fluid and energizer, average proppant placed per stage, average fluid pumped per stage, number of stages, and average factures spacing. First 6-months cumulative productions are utilized to characterize the after-stimulation performance of fractured horizontal wells. Results show that optimal clustering should be chosen by making a tradeoff between the Silhouette coefficient and the size of the clustered data sets. The optimal 2D and 3D clustering are 6 and 4 clusters obtained by AP and k-means algorithm, respectively. The dimensionality of input variable space can be greatly reduced by using the kernel PCA. The optimal number of principal components for ANN modeling is 7. The prediction ability of single layer neural network can be significantly improved by integrating DE algorithm into the ANN model. The determination coefficients (R-square) of validation and testing sets are increased by 31.0% and 23.8% for the optimal neural network. The developed data-driven technique provides a potential method to predict well after-stimulation performance. It is not only useful for evaluating the effects of stimulation parameters on well performance, but can also provide a reliable approach to predict well response towards different stimulation strategies in Montney tight formations.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".