Pump Flow Rate Estimation Using Motor Parameters and History
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining fluid flow rate through pumps, particularly centrifugal pumps, are costly and require hydraulic measurements, limiting their applicability and accuracy across various pump designs, including optimized impeller designs.
Innovation Solution
A computer-implemented method that calculates fluid flow rate using current and previous operational parameter values of the pump motor, such as rotational speed and electrical power, without requiring hydraulic measurements, employing a computational model that includes machine learning techniques like recurrent neural networks to accurately determine flow rates for diverse pump designs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If flow sensors are used to determine flow rate, then measurement accuracy is improved, but system cost increases
Solution Approach 1:
The patent replaces mechanical flow sensors with a computational model that uses electrical motor parameters (power, speed, current) to determine flow rate. This substitution eliminates the need for expensive hydraulic sensors by using already-available electrical measurements combined with machine learning algorithms to infer flow characteristics.
Solution Approach 2:
The patent introduces a computational model as an intermediary between electrical motor parameters and flow rate determination. Instead of directly measuring flow with sensors, the system uses the computational model to translate electrical parameters into flow rate estimates, serving as a virtual mediator that avoids the need for physical flow sensing.
2Measurement precision
If pressure measurements are taken at intake and discharge to determine flow rate, then measurement accuracy is improved, but device complexity increases
Solution Approach 1:
The patent extracts the flow rate determination capability from the hydraulic measurement system and relocates it to the electrical control domain. By taking out the need for pressure sensors and hydraulic measurements, the system achieves flow determination solely through electrical parameters already present in the motor control circuitry.
Solution Approach 2:
The patent replaces the mechanical hydraulic measurement system (pressure sensors, differential pressure transmitters) with an electrical computational approach. The machine learning model processes electrical signals to infer flow characteristics, substituting the entire mechanical measurement infrastructure with software-based computation.
3Productivity
If computational models use only current parameter values, then processing speed is improved, but measurement precision deteriorates
Solution Approach 1:
The patent applies preliminary action by using historical parameter values to pre-train and calibrate the computational model before actual flow determination. The model learns from past operational data (previous power, speed, current values) to establish accurate relationships, enabling faster real-time predictions with higher precision during operation.
Solution Approach 2:
The patent implements feedback by incorporating previous parameter values into the computational model. The system uses historical data as feedback to continuously refine and adjust the model's predictions, creating a closed-loop system where past performance informs future accuracy while maintaining real-time computational efficiency.
Data Source
AI summary
Computer-implemented methods are disclosed for determining a flow rate of fluid flow at a target time through a pump, such as a centrifugal pump, the pump being driven by a pump motor. An illustrative method includes, in some aspects: receiving at least one set of previous parameter values including a first previous parameter value indicative of a first operational parameter of the pump motor at a previous time, earlier than the target time, and a second previous parameter value indicative of a second operational parameter of the pump motor at the previous time, and receiving a current set of parameter values including a first current parameter value indicative of the first operational parameter of the pump motor at the target time and a second current parameter value indicative of the second operational parameter of the pump motor at the target time.