Pipe Leak Detection Using Frequency-Band Noise Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing leak detection methods in fluid media pipelines, such as water pipes, struggle to differentiate between leak noises and background disturbances effectively, especially when background noise levels exceed leak noise levels.
Innovation Solution
A method utilizing noise and frequency loggers with a processing unit that performs Fourier analysis to divide noise into frequency bands, sets individual threshold values for each band, and weights exceedances to distinguish leak noises from background noise, using FFT to analyze frequency components and adjust thresholds based on noise patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If noise level alone is used to detect leaks, then detection simplicity is maintained, but accuracy deteriorates due to inability to distinguish leak noise from background noise
Solution Approach 1:
The frequency spectrum is segmented into multiple frequency bands (e.g., 0-100 Hz, 100-200 Hz, 200-300 Hz, etc.), and threshold values are independently set for each band. This segmentation allows the system to distinguish leak noise characteristics from background noise patterns, improving detection accuracy while maintaining manageable complexity through standardized band definitions.
Solution Approach 2:
The detection approach transitions from using a single noise level parameter to evaluating multiple frequency band parameters simultaneously. By changing the evaluation parameter from monolithic noise level to distributed frequency spectrum, the system achieves better discrimination between leak and background noise sources.
2Measurement precision
If threshold values are set for each frequency band, then differentiation between leak and background noise is improved, but computational requirements increase
Solution Approach 1:
The frequency spectrum is divided into standardized frequency bands, and pre-defined threshold values are established for each band based on background noise characteristics. This segmentation with pre-computed thresholds reduces real-time computational burden while maintaining high differentiation capability between leak and background noise.
Solution Approach 2:
Threshold values for each frequency band are determined in advance during system setup or calibration phase, based on measurements of background noise levels. This preliminary action eliminates the need for complex real-time threshold calculation, reducing computational energy consumption during actual leak detection operations.
3Measurement precision
If high frequency components are emphasized, then leak detection sensitivity is improved, but susceptibility to interference from power grid noise increases
Solution Approach 1:
The frequency spectrum is segmented into multiple bands, allowing selective evaluation of high-frequency components (150-600 Hz) while separately monitoring lower frequency bands. This segmentation enables the system to recognize power grid interference patterns in specific bands and distinguish them from genuine leak noise signals.
Solution Approach 2:
The harmful interference from power grid noise at specific frequencies (50/60 Hz and 100/120 Hz) is converted into a beneficial detection feature by establishing threshold values that account for these known interference patterns. The system uses the presence and characteristics of these interference patterns to improve its ability to distinguish true leaks from false alarms.
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
Enhances the differentiation between leak and background noise, allowing for more accurate leak detection by focusing on frequency ranges specific to leak noises and reducing data volume through band reduction, thereby improving detection precision and reducing false alarms.
Implementation Method 1
The sensor is or preferably comprises a noise sensor and/or a piezo sensor and/or a piezo microphone
Implementation Method 2
a frequency analysis, in particular a Fourier analysis, preferably a Fast Fourier Transform, of the noise level is performed by means of the processing unit to determine a frequency spectrum
Implementation Method 3
a hydrophone for measuring noise directly at the water column
Data Source
Figure 1
AI summary
The invention relates to a method for detecting and/or analyzing a leak in a fluid media pipeline, in particular a water pipeline, wherein the method comprises the following steps: a) providing at least one noise and/or frequency logger, comprising a sensor, and a processing unit, b) attaching the noise and/or frequency logger to the pipeline or to fittings of the pipeline, c) measuring a noise at the pipeline by the sensor and determining a noise level (10) by means of the processing unit, c1) wherein a frequency analysis, in particular a Fourier analysis, of the noise level (10) is carried out by means of the processing unit to determine a frequency spectrum (20), c2) wherein the determined frequency spectrum (20) is divided into two or more frequency bands (21) by means of the processing unit, c3) wherein the processing unit, in particular individually,a threshold value (30) is defined for each frequency band and/or stored in the computing unit, which is greater than the maximum noise level (11) in that frequency band (21) when measured on the leak-free line, c4) wherein a leak in the line can be detected and/or analyzed by means of the computing unit based on the exceedance of one or more threshold values (30).