What question did this study set out to answer?

The aim is to improve accuracy and reliability of production forecasting using a novel hybrid modelling approach.

April 24, 2026EnergiesOpen Access

Physics-Guided Machine Learning with Flowing Material Balance Integration: A Novel Approach for Reliable Production Forecasting and Well Performance Analytics

View Full Paper

Ask AI

Bookmark

The aim is to improve accuracy and reliability of production forecasting using a novel hybrid modelling approach.
Developed hybrid models integrating flowing material balance into machine learning and physics-informed neural networks.
Created features from flowing material balance for neural network input.
Validated models using performance metrics like RMSE and R2 against field data.
Hybrid models show improved reliability in predicting production rates over the tested data period.
The proposed PiNN model effectively predicts well productivity indices through hyper-tuning.
PiNN models, while more reliable, demonstrated lower metric performance than conventional neural networks due to imposed constraints.

No takes yet. Share an insight, caveat, or question.

Hybrid modelling approach enhances forecasting accuracy in oil and gas reservoirs, suggesting new methodologies.

Motaei et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark