Predictive Model for Produced Water Element Concentration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting the concentration of elements in produced water from hydrocarbon wells are cumbersome and costly, especially in large reservoirs like shale gas production, where direct measurement of thousands of wells is impractical.
Innovation Solution
A computer-implemented machine-learning method that uses a predictive model trained on geoscience variables from other wells in the same hydrocarbon reservoir to estimate the concentration of elements like lithium, cobalt, or cadmium in produced water, allowing for accurate predictions without direct measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct well-wise measurement of element concentration is performed using ICP techniques, then measurement precision is improved, but productivity deteriorates due to the cumbersome nature of measuring thousands of wells
Solution Approach 1:
The patent creates a predictive model that copies the relationship between well characteristics and element concentrations learned from a training set of measured wells. This model then predicts concentrations for unmeasured wells without direct measurement, effectively creating a virtual copy of the measurement process that scales to thousands of wells while maintaining prediction accuracy
Solution Approach 2:
The patent performs preliminary measurements and model training on a representative training set of wells before actual production. This preliminary action creates a pre-trained predictive model that can quickly estimate concentrations for all other wells in the reservoir, avoiding the need to perform time-consuming ICP measurements on each individual well
2Reliability
If direct measurement of element concentration is performed for all wells, then reliability of concentration data is improved, but loss of time increases due to the extensive measurement process
Solution Approach 1:
The patent performs preliminary measurements on a training set of wells to establish the predictive model. Once trained, the model can rapidly predict concentrations for all other wells without requiring time-consuming ICP measurements, thus maintaining reliability while dramatically reducing the time required for concentration assessment across the entire reservoir
Solution Approach 2:
The predictive model copies the concentration patterns learned from measured wells and applies them to unmeasured wells. This allows the system to maintain reliable concentration estimates across all wells without repeating the time-intensive measurement process for each well
Data Source
AI summary
The invention notably relates to a computer-implemented method of machine-learning a predictive model configured for predicting a concentration of an element in produced water of a given well of hydrocarbon production in a hydrocarbon reservoir having wells of hydrocarbon production. The method comprises providing a dataset comprising values of one or more geoscience well-wise variables. Each value corresponds to a respective well of hydrocarbon production in the hydrocarbon reservoir other than the given well. Each value that corresponds to a respective well is associated to a respective ground truth value representing a concentration of the element in the respective well. The method further comprises learning the predictive model based on the dataset. This forms an improved solution for predicting a concentration of an element in produced water of a given well of hydrocarbon production in a hydrocarbon reservoir.