Ultrasonic Gas Flow Measurement Without Gas Composition Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ultrasonic flow measurement systems for gases, particularly natural gases, struggle to convert operating volumetric flow measurements into standard volumetric or mass flow measurements without precise knowledge of the gas composition or properties.
Innovation Solution
A computer-implemented method using machine learning to adapt parameters to the conditions at the measuring point, involving the creation and use of a dataset of test gases, determination of state variables, and regression analysis to approximate compressibility and sound velocity, allowing conversion of flow measurements without exact gas composition knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional ultrasonic flow measurement systems are used, then flow velocity can be measured, but accurate conversion to standard volumetric or mass flow measurements cannot be achieved without precise knowledge of gas composition
Solution Approach 1:
The patent introduces an intermediary machine learning model that acts as a mediator between the measured parameters (velocity, temperature, pressure) and the desired output (standard volumetric or mass flow). This model compensates for the missing gas composition information by learning complex relationships from training data, enabling accurate flow conversion without direct knowledge of the gas mixture composition.
Solution Approach 2:
The system transforms the measurement approach by changing from direct calculation requiring gas composition parameters to a machine learning-based parameter transformation. The model learns to map measured parameters (velocity, temperature, pressure) directly to standard flow conditions, effectively changing the measurement paradigm from physics-based calculation to data-driven prediction.
2Adaptability or versatility
If machine learning methods are introduced to adapt parameters automatically, then measurement accuracy improves under varying conditions, but device complexity increases
Solution Approach 1:
The machine learning model is trained in advance with extensive training data that covers various gas compositions and operating conditions. This preliminary training action allows the model to automatically adapt to different measuring conditions during operation without requiring real-time gas composition analysis or complex recalibration procedures, thereby achieving adaptability without proportional increases in operational complexity.
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
Enables accurate conversion of operating volumetric flow measurements to standard volumetric or mass flow measurements, even with unknown gas compositions, by automatically adapting to the prevailing conditions at the measuring point.
Implementation Method 1
The time-of-flight difference method utilizes the fact that the propagation speed of an ultrasonic signal depends on the flow velocity of the medium. Similar to a swimmer against the current, an ultrasonic signal travels slower against the flow direction than in the direction of flow.
Implementation Method 2
Ultrasonic flow meters measure the velocity of a flowing medium (gas, liquid) using acoustic waves.
Implementation Method 3
the total transit time is measured both upstream and downstream, from which the speed of sound in the fluid can then be calculated using the geometric parameters of the installation
Data Source
Figure 1
AI summary
The invention relates to a computer-implemented method for ultrasonic flow measurement and a device for carrying out the method. The object of the invention is to eliminate the disadvantages of the prior art and to provide a computer-implemented method for ultrasonic flow measurement of gases, in particular natural gases, by means of which an operating volumetric flow rate measurement can be converted into a standard volumetric or mass flow rate measurement without requiring knowledge of the exact gas composition or gas properties. This object is achieved by the features listed in the claims.