Two-Vane Pump Design Using Machine Learning for Wastewater

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current designs for two-vane pumps lack a systematic approach to achieve optimal efficiency and prevent clogging while handling wastewater, particularly in managing target heads and passable solid sizes, leading to inefficiencies and potential damage.

Innovation Solution

A machine learning-based design method for two-vane pumps that sets objective functions for head, efficiency, and passable solid volume, using radial basis neural networks and genetic algorithms to optimize impeller and volute design variables, ensuring high efficiency and preventing clogging.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a vortex pump is designed to prevent clogging by shortening the impeller length, then clogging is prevented, but pump efficiency decreases significantly

Engineering Contradiction:
Improveclogging preventionVSAvoidpump efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The impeller is divided into two separate vanes instead of a single continuous impeller. This segmentation allows each vane to be optimized independently for both clogging prevention and efficiency, with the symmetrical arrangement ensuring balanced fluid flow while maintaining wide passage paths that prevent clogging.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent employs symmetrical asymmetry by using two identical vanes arranged symmetrically around the rotation center. This creates a balanced structure that prevents clogging through wide passages while maintaining efficient fluid flow patterns, resolving the contradiction between clogging prevention and pump efficiency.

Inventive Principle:
Principle #4Asymmetry

2Productivity

If a single channel pump is used to achieve high pump efficiency, then pump efficiency increases, but vibrations increase significantly due to asymmetric impeller structure

Engineering Contradiction:
Improvepump efficiencyVSAvoidvibration distribution
Core Design Contradiction:
ProductivityVSStability of the object's composition

Solution Approach 1:

The patent uses symmetrical asymmetry with two identical vanes arranged symmetrically around the rotation center. This creates a balanced structure that maintains high pump efficiency while significantly reducing vibrations compared to single-channel pumps with asymmetric impellers. The symmetrical arrangement ensures uniform fluid force distribution.

Inventive Principle:
Principle #4Asymmetry

Solution Approach 2:

By segmenting the impeller into two separate vanes, the patent achieves a balanced structure that distributes fluid forces uniformly, reducing vibrations while maintaining the efficient flow paths characteristic of single-channel pump designs.

Inventive Principle:
Principle #1Segmentation

3Ease of manufacture

If conventional impeller design is used without systematic optimization, then manufacturing is simpler, but achieving target head and passable solid size requires extensive trial and error

Engineering Contradiction:
Improveimpeller structure simplicityVSAvoidtrial and error time
Core Design Contradiction:
Ease of manufactureVSLoss of time

Solution Approach 1:

The patent systematically optimizes key parameters including the angle between vanes, vane curvature radius, and spacing between vanes. By establishing design formulas that relate these parameters to target head and passable solid size, the patent eliminates extensive trial and error while maintaining manufacturable simple structures.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent performs preliminary design calculations using established formulas to determine optimal vane angles, curvature radii, and spacing before manufacturing. This preliminary optimization action eliminates the need for extensive trial and error during the design process, saving time while maintaining manufacturing simplicity.

Inventive Principle:
Principle #10Preliminary action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The method effectively enhances pump efficiency, prevents clogging, and allows for the design of two-vane pumps with desired specifications, improving flow rates and reducing vibrations, thereby increasing operational reliability.

Implementation Method 1

a pair of impellers configured to rotate inside the volute and formed in a symmetrical streamline shape with respect to a rotation center to guide a flow of a fluid introduced into the volute

Methodology Applied
Scientific EffectCentrifugal force: Centrifugal Force

Implementation Method 2

a two-vane pump for wastewater including a circular volute having one inlet and one outlet, and a pair of impellers configured to rotate inside the volute

Methodology Applied
Scientific EffectHydraulic pump mechanism: Pump

Data Source

PatentUS11953023B2Two-vane pump and design method of two-vane pump for wastewater using machine learning
Publication Date: 2024.04.09 KOREA INSTITUTE OF INDUSTRIAL TECHNOLOGY
  • US11953023B2 patent drawing
  • US11953023B2 patent drawing
  • US11953023B2 patent drawing

AI summary

The present disclosure relates to a two-vane pump for wastewater and a design method of a two-vane pump for wastewater using machine learning, and more particularly, a design method of a two-vane pump using machine learning capable of having efficiency of a target head and performing optimal design for sizes of solids that can pass through and a two-vane pump for wastewater according to the machine learning. According to the present disclosure, there is provided a design method of two-vane pump for wastewater using machine learning, including: a) setting an objective function; b) setting design variables of the impeller and volute for deriving the set objective function and a range of each design variable; c) deriving a plurality of experimental points including values of the design variables within the range of the design variable; d) generating an input value by calculating the value of the objective function through numerical analysis of each of the derived experimental points; e) constructing a surrogate model through machine learning for the input value; and f) deriving an optimal design of the two-vane pump for wastewater from the constructed surrogate model.