Predictive Model for Well Completion Design Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Customizing well completion designs for unconventional subsurface reservoirs is challenging due to local variability in petrophysical properties, making it difficult to effectively stimulate these reservoirs with a single design, as there are numerous combinations of completion designs and subsurface conditions.
Innovation Solution
A computer-implemented method using multivariate imputed data to predict well production trends by generating a predictive model from input data that includes subsurface and well completion data, decorrelating and correlating the data to fill gaps in original multivariate data, and applying this imputed data to visualize production trends against various completion designs and conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single well completion design is used, then the device complexity is reduced, but the effectiveness of stimulating unconventional reservoir rocks across the field deteriorates due to local variability in petrophysical properties
Solution Approach 1:
The patent applies local quality by customizing well completion designs to match specific subsurface conditions. The system divides the field into different zones based on petrophysical properties and assigns optimized completion parameters (lateral well length, proppant amount, frack water volume, fracture cluster number) to each zone, ensuring that each local area receives a design tailored to its specific characteristics rather than a uniform approach
2Reliability
If well completion designs are customized per subsurface conditions, then the reservoir stimulation effectiveness is improved, but the difficulty of determining optimal designs increases due to the large number of combinations of completion designs and subsurface conditions
Solution Approach 1:
The patent uses copying by creating a predictive model that replicates the complex relationship between completion designs and production outcomes. Instead of performing time-consuming reservoir simulations for each design scenario, the system builds a predictive model from historical data that can quickly copy and evaluate the performance of different completion designs under various subsurface conditions, significantly reducing the computational burden
Solution Approach 2:
The patent applies preliminary action by pre-processing the input data through decorrelation transformations before applying the predictive model. The system performs data normalization and feature engineering in advance, organizing the complex multi-parameter data into a structured format that accelerates the prediction process and makes it easier to evaluate multiple design combinations
3Measurement precision
If traditional reservoir simulations are used to determine optimal well completion designs, then the accuracy of production predictions is improved, but the time required for analysis increases significantly
Solution Approach 1:
The patent replaces time-consuming reservoir simulations with a predictive model that copies the essential relationships from historical data. The model is trained on existing production data and completion parameters, creating a simplified representation that can rapidly predict production outcomes without requiring full-scale numerical reservoir simulations, thus maintaining accuracy while dramatically reducing computation time
4Measurement precision
If more data is collected from multiple wells to improve predictive accuracy, then the measurement precision of production trends is improved, but the complexity of processing and analyzing the multivariate data increases
Solution Approach 1:
The patent extracts the essential relationships from complex multivariate data by using the predictive model to isolate the most important factors affecting production. The system separates the signal from noise in the historical data, extracting key patterns and relationships between completion parameters and production outcomes, thereby simplifying the analysis of multi-well data while maintaining predictive accuracy
Data Source
AI summary
Example computer-implemented methods, media, and systems for determining well production trend using subsurface condition data, well completion data, and well production data are disclosed. One example computer-implemented method includes obtaining first data associated with multiple wells, where the first data includes input data and well production data, and the input data includes subsurface condition data and well completion data. A first transformation decorrelates the input data into the second data. Multiple random numbers are generated using the second data. A second transformation correlates the multiple random numbers into imputed data of the input data, where the second transformation includes an inverse transformation of the first transformation. A predictive model between the well production data and the input data is applied to the imputed data of the input data to generate imputed data of the well production data and to predict well production trend of the multiple wells.