Gas Leak Flow Rate Estimation via Machine Learning Regression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gas leak monitoring systems are ineffective in quickly determining the instantaneous gas flow rate, which is crucial for assessing health and environmental risks during accidental releases of hazardous gases like H2S and CO2.
Innovation Solution
A system utilizing a regression model implemented by a machine-learning subsystem, specifically Gaussian Process Regression (GPR) or deep neural networks (DNN), that processes concentration profiles from laser-based monitoring systems and field tests to estimate gas leak flow rates in real-time, allowing for immediate attenuation actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional gas leak monitoring systems are used, then gas concentration can be detected, but the instantaneous gas flow rate cannot be quickly determined
Solution Approach 1:
The patent introduces an intermediary machine learning model that mediates between the measured gas concentration data and the desired flow rate calculation. The model is trained on simulation data to learn the complex relationship between concentration profiles and flow rates, enabling quick estimation without direct measurement. This intermediary model resolves the contradiction by providing fast flow rate estimates based on concentration data.
Solution Approach 2:
The patent performs preliminary action by pre-training the machine learning model using extensive simulation data before actual gas leak detection. The simulation data encompasses various leak scenarios, wind conditions, and terrain features. This pre-training prepares the model to quickly estimate flow rates during actual events without requiring time-consuming real-time calculations, thus resolving the time delay contradiction.
2Measurement precision
If detailed simulation data is used to train the model, then estimation accuracy improves, but data processing complexity increases
Solution Approach 1:
The patent creates a simplified copy of the complex physical system through machine learning. Instead of running complex simulations in real-time, the system uses pre-trained model weights that capture the essential relationships from simulation data. This copying approach maintains high accuracy while reducing real-time computational complexity, as the model inference is much faster than running full simulations.
3Speed
If real-time flow rate estimation is implemented, then response time to gas leaks improves, but computational resources required increase
Solution Approach 1:
The patent performs computationally intensive work in advance by training the machine learning model on extensive simulation data before deployment. Once trained, the model requires minimal computational resources for real-time inference during actual gas leak events. This preliminary action shifts the energy consumption from real-time operation to offline training, enabling fast response with low real-time energy usage.
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 quick and accurate estimation of gas flow rates, facilitating timely and effective responses to gas leaks, thereby reducing health and environmental hazards.
Implementation Method 1
Modern monitoring techniques exist that are based on laser beam absorption by the contaminant's molecules
Data Source
AI summary
An estimated gas leak flow rate can be determined using a teaching set of concentration profiles, a regression model implemented by a machine-learning subsystem, and a subset of attributes measured within an environment. The teaching set of concentration profiles can include gas flow rates associated with relevant attributes. The regression model can be transformed into a gas leak flow regression model via the machine-learning subsystem using the teaching set. The subset of attributes measured within the environment can be applied to the gas leak flow regression model to determine other attributes absent from the subset of attributes and an estimated gas flow rate for the environment. A gas leak attenuation action can be performed in response to the estimated gas flow rate.