Rate Control Algorithm Optimizing Fuel Distribution

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Optimizing fuel distribution in a fleet of engine-driven pumps during fracturing operations is challenging due to the inability of manual intervention to achieve optimal substitution rates, especially when a mix of single and dual fuel engines are used, as existing algorithmic models do not consider unique parameters of individual pumps such as air filter conditions, engine hours, and pump hours.

Innovation Solution

A control unit and rate control algorithm with a trained algorithmic model process operation variables to determine optimal fuel distribution rates, considering historical data and unique parameters of each engine-driven pump, generating rate control variables that are distributed based on a selected control unit identifier to maximize fuel substitution and minimize greenhouse gas emissions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual intervention is used to operate pumps, then operator experience can be applied, but optimal fuel substitution cannot be achieved across a fleet of pumps

Engineering Contradiction:

Improvemanual operationVSAvoidfuel substitution optimization

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The patent replaces manual mechanical operation with an automated computer-based rate control algorithm that uses trained algorithmic models to determine optimal fuel distribution rates across the fleet, eliminating the need for manual intervention while achieving superior optimization results

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service operation where the automated algorithm independently monitors pump parameters, adjusts fuel distribution rates, and optimizes natural gas substitution without requiring continuous manual input or intervention from operators

Inventive Principle:

Principle #25Self-service

2Extent of automation

If existing algorithmic models are used, then automation is provided, but unique parameters of individual pumps are not considered

Engineering Contradiction:

Improverate distribution automationVSAvoidindividual pump parameter consideration

Core Design Contradiction:

Extent of automationVSMeasurement precision

Solution Approach 1:

The patent applies local quality by training separate algorithmic models for each individual pump based on its unique parameters and characteristics, allowing each pump to receive customized rate recommendations that account for its specific air filter conditions, engine hours, pump hours, and other individual traits rather than applying a uniform model to all pumps

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system segments the fleet management problem by creating distinct trained models for each pump unit, dividing the overall optimization task into individual pump-specific sub-problems that can be solved independently and then aggregated to achieve fleet-wide optimization

Inventive Principle:

Principle #1Segmentation

3Adaptability or versatility

If a fleet of mixed single and dual fuel pumps is used, then operational flexibility is increased, but manual rate distribution becomes very challenging

Engineering Contradiction:

Improvefuel type flexibilityVSAvoidrate distribution complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a universal rate control algorithm that can handle both single-fuel and dual-fuel pump types through a single automated system, eliminating the need for separate manual management approaches for different fuel configurations and providing consistent optimization across the entire fleet regardless of fuel type

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS11598273B2Rate control algorithm having a trained algorithmic model for optimizing fuel distribution

Publication Date: 2023.03.07 HALLIBURTON ENERGY SERVICES INC

AI summary

A system comprising engine driven pumps, control units, a rate control algorithm, and a trained algorithmic model. The control units store operation variables for the engine driven pumps. The rate control algorithm includes an instruction set to read the operation variables for the engine driven pumps. The instruction set comprises a trained algorithmic model with a parameter space based on historical operation variables. The trained algorithmic model determines an optimal distribution rate based on an objective. The instruction set generates rate control variables based on the determined optimal distribution rate. Each rate control variable comprises a selected control unit identifier and a rate value. The instruction set distributes each rate control variable based on the selected control unit identifier. The trained algorithmic model determine the optimal distribution rate using an objective that defines a desired mixture between a first fuel and a second fuel.