Advance artificial time series forecasting model for oil production using neuro fuzzy-based slime mould algorithm

Area of Science:

• Petroleum Engineering
• Artificial Intelligence
• Time Series Analysis

Background:

• Effective management of petroleum reservoirs relies heavily on accurate oil production forecasting.
• Existing forecasting models may face limitations such as local optima during optimization.

Purpose of the Study:

• To develop and evaluate a novel time series forecasting model for oil production.
• To improve the adaptive neuro-fuzzy inference system (ANFIS) using an optimization algorithm.

Main Methods:

• A new adaptive neuro-fuzzy inference system (ANFIS) model was developed.
• The slime mould algorithm (SMA) was enhanced with opposition-based learning (OLB), creating the ANFIS-SMAOLB model.
• The model was validated using real-world oil production data from Masila (Yemen) and Tahe (China) oilfields.

Main Results:

• The ANFIS-SMAOLB model demonstrated high accuracy and efficiency in time series forecasting.
• Extensive comparisons showed significant performance improvements over other methods.
• The model effectively addressed limitations of the original SMA, such as avoiding local optima.

Conclusions:

• The developed ANFIS-SMAOLB model is a highly capable tool for oil production forecasting.
• This enhanced ANFIS model offers significant performance advantages for petroleum reservoir management.
• The integration of OLB with SMA provides a robust optimization technique for time series forecasting.