Gas Network Leak And Obstruction Detection With Black-Box Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to effectively detect and quantify leaks and obstructions in complex gas networks, particularly those between sources and consumers, due to assumptions based on long, straight pipelines and do not account for complex network topologies.
Innovation Solution
A method utilizing a mathematical model trained with sensor data from adjustable relief and throttle valves to simulate leaks and obstructions, allowing for simultaneous detection and quantification without requiring knowledge of the exact network topology, using a black-box model to analyze sensor data and generate alarms for irregularities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional leak detection methods based on long straight pipeline assumptions are used, then the detection method is simple, but it cannot detect leaks and obstructions in complex gas networks with multiple consumers and compressor plants
Solution Approach 1:
The patent creates a virtual copy of the gas network through mathematical modeling. A black-box model replicates the network's behavior by establishing relationships between sensor measurements at different locations, allowing leak and obstruction detection in complex topologies without requiring detailed knowledge of the actual network structure. This virtual copy enables detection capabilities previously unavailable for complex networks.
Solution Approach 2:
The patent replaces traditional mechanical/physical leak detection methods with a data-driven mathematical approach. Instead of using physical sensors that require specific pipeline geometries, the system uses sensor data from existing monitoring points combined with machine learning models to detect leaks and obstructions, substituting mechanical detection principles with computational analysis.
2Reliability
If existing leak detection methods are used, then the implementation is straightforward, but they cannot simultaneously detect both leaks and obstructions in the gas network
Solution Approach 1:
The patent develops a universal detection system that can identify both leaks and obstructions using the same mathematical framework and sensor data. The black-box model is designed to handle multiple failure modes simultaneously by analyzing patterns in sensor measurements, allowing a single system to perform multiple detection functions that previously required separate methods.
Solution Approach 2:
The system uses feedback from multiple sensor locations throughout the network to detect and characterize leaks and obstructions. By continuously monitoring pressure and flow data at various points and comparing it against the mathematical model predictions, the system can identify anomalies indicating leaks or obstructions and provide feedback for location and severity determination.
3Measurement precision
If the gas network is shut down for leak and obstruction detection, then accurate detection can be performed, but network productivity is reduced
Solution Approach 1:
The patent enables continuous leak and obstruction detection while the gas network remains operational. The mathematical model processes sensor data from the running network, allowing detection activities to proceed without interrupting gas flow or network operations. This maintains continuous useful action of both detection and network operation simultaneously.
Data Source
Figure 1
Figure 2
AI summary
Method for the simultaneous detection, localization and quantification of leaks (13a) and obstructions (13b) in a gas network (1) under pressure or vacuum; the gas network (1) comprising: - one or more sources (6) of compressed gas or vacuum; - one or more consumers (7) or consumer areas of compressed gas or vacuum applications; - pipelines (5) or a network (4) of pipelines (5) to transport the compressed gas or vacuum from the sources (6) to the consumers (7), consumer areas or applications; - a plurality of sensors (9a, 9b, 9d) providing one or more physical parameters of the gas at different times and locations within the gas network (1); characterized in that the gas network (1) is further provided with a number of controllable or adjustable relief valves (10a), a number of controllable or adjustable throttle valves (10b) and possibly one or a plurality of sensors (9c) capable of monitoring the status or state of the relief valves (10a) and/or throttle valves (10b) and that the method comprises the following steps: - a training phase (16), in which a mathematical model is established between the measurements of a first group of sensors (9a, 9b, 9c, 9d) and a second group of sensors (9a, 9b, 9c, 9d), based on different measurements of these sensors (9a, 9b, 9c, 9d), wherein the controllable or adjustable relief valves (10a) and throttle valves (10b) are controlled in a predetermined sequence and according to well-designed scenarios to generate leaks (13a) and obstructions (13b) respectively; - an operational phase (17), in which the mathematical model established between the measurements of the first group of sensors (9a, 9b, 9c, 9d) and the second group of sensors (9a, 9b, 9c, 9d) is used to detect, locate and quantify leaks (13a) and obstructions (13b) in the gas network; wherein the operational phase (17) comprises the following steps: - controlling, if necessary, the relief valves and the throttle valves in a predetermined order and according to well-designed scenarios; - reading out the first group of sensors (9a, 9b, 9c, 9d); - based on these readout measurements, calculating or determining the value of the second group of sensors (9a, 9b, 9c, 9d) with the help of the mathematical model; - comparing the calculated or determined values of the second group of sensors (9a, 9b, 9c, 9d) with the read values of the second group of sensors (9a, 9b, 9c, 9d) and determining the difference between them; - determining whether there is a leak (13a) and/or an obstruction (13b) in the gas network on the basis of the aforementioned difference and any of its derivatives; - generating an alarm if a leak (13a) or obstruction (13b) is detected and/or determining the location of the leak (13a) and/or obstruction (13b) and/or determining the flow rate of the leak (13a) and/or the degree of obstruction of the obstruction (13b) and/or generating the leakage and/or obstruction cost.