Pump Control Parameters via ML Application Classification

Solution Overview

Problem

Existing pump control methods require manual user input and knowledge to adjust control parameters, often leading to suboptimal operation, malfunctions, and reduced pump lifespan due to improper settings.

Innovation Solution

A computer-implemented method using a trained machine-learning model to automatically determine application-specific control parameters for a pump by monitoring operational variables, allowing for adaptive control based on the identified application type.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If manual adjustment of control parameters is performed, then the pump can be adapted to specific applications, but the process is time-consuming and requires user knowledge

Engineering Contradiction:

Improveadaptation to specific applicationVSAvoidadjustment time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The pump system automatically determines its own application type by monitoring operational variables and applying a machine-learning model, eliminating the need for user intervention. The system self-configures control parameters based on the determined application type, enabling autonomous adaptation without requiring user knowledge or time for manual adjustment.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The manual mechanical adjustment process is replaced by an automated electronic system using machine-learning algorithms. The system substitutes human decision-making with computational analysis of operational variables, automatically selecting appropriate control parameters through software-based classification rather than manual configuration.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If manual adjustment of control parameters is performed, then the pump can be adapted to specific applications, but it requires user knowledge and expertise

Engineering Contradiction:

Improveadaptation to specific applicationVSAvoidease of configuration

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The pump system automatically determines its own application type by monitoring operational variables and applying a machine-learning model, eliminating the need for user intervention. The system self-configures control parameters based on the determined application type, enabling autonomous adaptation without requiring user knowledge or time for manual adjustment.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The manual mechanical adjustment process is replaced by an automated electronic system using machine-learning algorithms. The system substitutes human decision-making with computational analysis of operational variables, automatically selecting appropriate control parameters through software-based classification rather than manual configuration.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If factory default settings are used, then the pump can be installed quickly, but the operation becomes suboptimal leading to reduced performance and lifespan

Engineering Contradiction:

Improveinstallation speedVSAvoidpump performance

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system performs preliminary monitoring of operational variables during the initial operation phase to determine the application type. Based on this determination, the control parameters are automatically adjusted to optimal settings for the specific application, ensuring both quick installation and optimal performance without requiring post-installation manual configuration.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors operational variables and uses this feedback to automatically adjust control parameters. The machine-learning model processes the monitored data and dynamically optimizes pump operation based on the determined application type, ensuring optimal performance while maintaining installation efficiency.

Inventive Principle:

Principle #23Feedback

Data Source

PatentEP3957863A1Method and system for controling a pump

Publication Date: 2022.02.23 GRUNDFOS HLDG

Figure 1~2
Figure 3
Figure 4

AI summary

A computer-implemented method for determining one or more application-specific control parameters of a pump operating in a system for moving a fluid, the method comprising: monitoring one or more operational variables of the pump during operation of the pump in said system; applying a trained machine-learning model to automatically determine an application type from the monitored operational variables, the application type representing a type of system the pump operates in and/or representing a type of operation performed by the pump when operating in said system, each application type being associated with a respective set of one or more application-specific control parameters; controlling the pump based on the set of one or more application-specific control parameters associated with the determined application type.