Vehicle Braking Noise Classification Using AI Spectrogram Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current noise analyzers for vehicle braking systems are ineffective in reliably detecting and classifying various noise types due to their limitations in handling complex friction-induced vibrations, leading to false positives and negatives, and fail to distinguish between noise classes and anomalies, such as low-frequency noise and higher-order harmonics, which affects the accuracy of noise recognition and classification.
Innovation Solution
A method utilizing artificial intelligence and machine learning algorithms, including neural networks, to analyze digital audio data from braking systems, filtering out low-frequency noise, and generating spectrograms to identify and classify noise events into specific categories like squeals, chirp/wirebrush, artifacts, and anomalies, with the aid of tagging and segmentation techniques to improve detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard noise analyzers using Fourier transform-based spectral processing methods are used to identify noise events, then tonal noise (squeal) detection is improved, but detection of other noise types (collision noise, chirp/wirebrush, artifacts) deteriorates
Solution Approach 1:
The patent segments the noise analysis process into multiple specialized modules: a spectral analysis module for tonal noise detection using Fourier transforms, and a time-domain analysis module for transient noise detection using wavelet transforms and energy-based features. This segmentation allows each module to specialize in specific noise types, resolving the contradiction between squeal detection precision and overall noise type versatility.
Solution Approach 2:
The patent creates a universal noise analysis system that integrates multiple analysis methods (Fourier transform, wavelet transform, time-domain energy analysis) into a single platform. The system can adaptively select and combine different analysis techniques based on the characteristics of the input signal, enabling it to detect and classify multiple noise types including squeal, collision noise, chirp/wirebrush, and artifacts with high accuracy.
2Speed
If noise analyzers rely on amplitude criteria and spectral processing methods, then detection speed is improved, but reliability in noisy environments deteriorates
Solution Approach 1:
The patent introduces wavelet transforms as an intermediary analysis method that bridges the gap between fast spectral analysis and reliable transient detection. Wavelet transforms provide time-frequency localization that allows the system to quickly identify transient noise events while maintaining reliability in noisy environments by analyzing both temporal and spectral characteristics simultaneously.
Solution Approach 2:
The patent dynamically adjusts analysis parameters such as window size, overlap percentage, and threshold levels based on the characteristics of the input signal and environmental noise conditions. This adaptive parameter adjustment allows the system to maintain high detection speed while improving reliability in varying noise environments by optimizing the balance between sensitivity and false alarm rates.
3Measurement precision
If engineers manually classify noise based on spectrograms, then discrimination accuracy is improved, but analysis time increases
Solution Approach 1:
The patent implements automated noise classification algorithms that extract features from raw audio signals and automatically classify noise types without requiring manual spectrogram analysis. The system uses machine learning classifiers trained on labeled noise data to automatically distinguish between different noise types, maintaining high classification accuracy while eliminating the time-consuming manual analysis process.
Solution Approach 2:
The patent replaces the manual mechanical process of spectrogram visualization and human interpretation with automated computational algorithms. The system automatically performs time-frequency analysis, extracts relevant features, and classifies noise types using computational models, substituting human expertise with algorithmic processing that achieves comparable or superior accuracy at much higher speeds.
Data Source
AI summary
A method for identifying and characterizing noises generated by a vehicle braking system is described. The method first comprises the steps of detecting noises generated by a vehicle braking system under dynamic operating conditions and generating digital audio data representative of the detected noise. The method then provides analyzing the aforesaid digital audio data by a noise analyzer, to identify potential squeal events and respective likely squeal frequencies, and generating squeal frequency information indicative of the squeal frequencies of the identified potential squeal events. The method then comprises the steps of filtering the aforesaid digital audio data by means of high-pass filtering to eliminate spectral components at frequencies lower than a filtering frequency, to generate filtered digital audio data; and generating, based on the filtered digital audio data, a respective spectrogram, which represents, in graphical form, information present in the filtered digital audio data, comprising the sound signal intensity, as a function of time and frequency. The method then involves providing the aforesaid spectrogram and the aforesaid squeal frequency information to a trained algorithm, wherein the algorithm was trained using artificial intelligence and/or machine learning techniques. The method also provides identifying noise events, by the trained algorithm, based on the above spectrogram and squeal frequency information, classifying the identified noise events and finally providing information about the identified noise events, each characterized by the respective category. The aforesaid classification step involves a classification according to at least the following categories: a first category comprising noises to be detected generated by the characteristic dynamic operation of the braking system; and a second category comprising abnormal noises, generated by operational or test anomalies.