Well Model Flow Tables Without Database Middleware
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine learning models in the oil and gas industry require large databases for storing millions of records, leading to expensive and time-consuming middleware requirements, and integration logic for physical tables, which hampers efficient data processing and model updates.
Innovation Solution
Generating well model flow tables using object-oriented object instances and result objects, which are encapsulated in a data object to directly pass to a machine learning model without database tables, leveraging Python objects to eliminate middleware and integration logic, and automating model updates with physical models and dedicated hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional database tables are used to store flow table data, then data can be stored and accessed, but storage requirements become very large (millions of records) and middleware complexity increases
Solution Approach 1:
The patent extracts the essential data representation from traditional database tables and implements it using lightweight Python objects (pandas DataFrames). This removes the need for complex database middleware while retaining the ability to store and manipulate flow table data efficiently, directly reducing both storage requirements and system complexity
Solution Approach 2:
The patent creates simplified copies of database table functionality using Python data structures. Instead of relying on full database systems, it uses pandas DataFrames that replicate tabular data storage and manipulation capabilities with minimal overhead, eliminating the need for expensive database middleware
2Reliability
If physical models with integration logic are used, then accurate well performance modeling is achieved, but integration logic complexity and processing time increase
Solution Approach 1:
The patent replaces traditional mechanical integration logic (complex computational workflows) with Python-based object-oriented implementations. This substitution maintains modeling accuracy while significantly reducing processing time through optimized Python code execution and vectorized operations in pandas
Solution Approach 2:
The patent changes the implementation parameters of physical models from traditional software integration to Python-based object instances. This parameter change allows the same physical modeling logic to execute faster while maintaining accuracy, as Python objects can be instantiated and manipulated more efficiently than traditional integration architectures
3Quantity of substance
If large databases are used for machine learning model training, then comprehensive data is available, but data processing latency increases
Solution Approach 1:
The patent extracts data from large database systems and loads it into memory-efficient pandas DataFrames. This extraction eliminates the need for continuous database connectivity and reduces data processing latency by working with in-memory data structures that can be rapidly accessed and manipulated during machine learning model training
Data Source
AI summary
Systems and methods include a computer-implemented method for well models. Historical well data for past operations of a well is received. An expanded minimum/maximum range of values for the historical well data is generated for parameters used in a well model. Object-oriented object instances are generated using the historical well data and the expanded range. Each object instance is an object that includes a specific combination of parameters within the expanded range. The well model is executed using the instances, generating result objects, each representing a result of the well model. A hybrid well model variant of the well model is generated using the object instances and result objects to model combinations of the object instances and the result objects. A data object encapsulating the object instances and result objects is passed by an object-oriented application to a machine learning model in an absence of database tables.