ESP Sizing Recommendation Engine for Wellbore-Specific Pump Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electrically submersible pumps (ESPs) used in wellbore applications face variability in reliability and affordability due to differing environmental conditions and operator preferences, necessitating a system that can be tailored to specific wellbore environments and operational plans.
Innovation Solution
A machine learning-based approach is employed to identify and select ESP components that match job-specific characteristics, environmental conditions, and operator preferences, using a neural network to optimize pump design and operation based on reliability and cost metrics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single ESP design is used for multiple wellbore applications, then device complexity is reduced, but reliability decreases due to mismatch with specific environmental conditions and operational plans
Solution Approach 1:
The patent implements a dynamic ESP sizing and configuration system that adapts to different wellbore environments through machine learning algorithms. The system dynamically selects pump specifications, motor parameters, and operational settings based on real-time wellbore characteristics, replacing static one-size-fits-all designs with adaptive, condition-based configurations.
Solution Approach 2:
The system changes critical parameters such as pump discharge capacity, motor power rating, and rotational speed based on wellbore-specific conditions including temperature, pressure, fluid viscosity, and production goals. This parameter optimization ensures each ESP configuration is tailored to maximize reliability for its specific application environment.
2Reliability
If ESP components are customized for specific wellbore conditions, then reliability improves, but manufacturing cost increases
Solution Approach 1:
The machine learning model performs preliminary analysis of wellbore data before ESP manufacturing begins. By pre-calculating optimal specifications and identifying potential reliability risks, the system enables manufacturers to build customized pumps with predetermined configurations, reducing on-site adjustments and ensuring reliability while controlling manufacturing costs through efficient design-phase optimization.
3Productivity
If ESP systems are optimized for maximum production capacity, then productivity increases, but reliability decreases due to higher operational stress and reduced affordability
Solution Approach 1:
The system applies partial optimization by selecting ESP configurations that provide sufficient (but not maximum) production capacity for each wellbore application. The machine learning algorithms calculate the minimum reliable configuration needed, avoiding over-engineering that would increase operational stress and cost, while still meeting production goals through optimized parameter selection rather than maximum capacity designs.
Data Source
AI summary
Aspects of the subject technology relate to systems, methods, and computer-readable media for building electrically submersible pump (ESP) systems. ESP systems include parts that must be compatible with a wellbore environment for them to operate as desired. Parts that operate well when pumping oil may not operate well when pumping geothermal water. Parts of an ESP system must be able to fit into the wellbore and operate according to the expectations of a wellbore operator. This means that parts used to build an ESP system must be selected such that the ESP system may be operated according to an operational plan. Once an ESP system is designed and built, it may be operated in a manner that is consistent with an operational plan. Techniques of the present disclosure include machine learning processes that identify how best to design, build, and operate an ESP system based on a set of characteristics.