Integrating Physics and Machine Learning for Unified Seismic Forward Modeling and Inversion of Reservoir Properties
**# Python Code Modules for Petroleum and Natural Gas Reservoir Analysis** This file contains Python code modules developed for two types of reservoir systems: liquid petroleum and natural gas. Each type is categorized into four typical cases based on porosity level (high/low) and reservoir fluid properties (hydrocarbon-bearing/water-bearing). The code execution specifications are as follows: **### Basic Analysis Modules (Items 1-6, 14-20)** 1.Used for exploring reservoir parameter distribution patterns 2. Each scenario only needs to be executed once **### Scenario Simulation Modules (Items 7-13)** 1. Includes differential simulations for por
Resource information
- Format
- zip
- Original repository
- 17089793
- DOI
- 10.5281/zenodo.17089793
- Credit basis
- Creators in Zenodo record metadata
- Access
- Open files available from the original Zenodo deposit
- Last checked
- Nov 5, 2026
- License
- cc-by-4.0
- Last record update
- Nov 5, 2026
About this resource
**# Python Code Modules for Petroleum and Natural Gas Reservoir Analysis** This file contains Python code modules developed for two types of reservoir systems: liquid petroleum and natural gas. Each type is categorized into four typical cases based on porosity level (high/low) and reservoir fluid properties (hydrocarbon-bearing/water-bearing). The code execution specifications are as follows: **### Basic Analysis Modules (Items 1-6, 14-20)** 1.Used for exploring reservoir parameter distribution patterns 2. Each scenario only needs to be executed once **### Scenario Simulation Modules (Items 7-13)** 1. Includes differential simulations for porosity-fluid combinations 2. Must be executed separately for each of the four case types, for a total of four executions **### Noise Processing Module (Item 4)** 1. Implements random noise injection with SNR=40dB parameter for 100 iterations 2. To avoid interference from randomness, the current version is fixed to execute a single noise generation **##Sensitivity Analysis of Classical AVO Inversion (Aki–Richards Approximation)** The following Python scripts analyze two key factors affecting the accuracy of classical AVO inversion methods (using the Aki–Richards approximation as an example): 1. The influence of variations in the initial model on the inversion errors of reservoir elastic parameters. (Aki_richards_inversion_mc_analysis) 2. The impact of seismic AVO data noise level, quantified by the signal-to-noise ratio (SNR), on inversion accuracy. (Aki_richards_inversion_snr_sensitivity) **## Environment Configuration Requirements** 1. Path Settings: Absolute paths in the code must be replaced with valid local paths 2. Development Stack: Python 3.12 + PyCharm 2024.1.6 Professional Edition + Anaconda conda 24.11.3 3. Minor result fluctuations may occur due to version differences **## Technical Support** For technical inquiries: ;