Electric Vehicle Sound Quality Index Prediction via Noise Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for evaluating sound quality indexes in electric vehicles rely heavily on developer analysis and experience, leading to limitations in qualitative assessments and inaccuracies in predicting noise, vibration, harshness (NVH) performance.
Innovation Solution
A method is developed to predict a sound quality index by separating high-frequency whine noise, background noise, and overall noise using interior noise data, motor revolutions per minute (rpm) data, and speed data of an electric vehicle, and applying regression analysis to create a predictive model with high correlation to subjective evaluations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional interior noise data evaluation methods are used, then the evaluation process is simple, but the prediction accuracy of sound quality index is low
Solution Approach 1:
The patent segments the interior noise into distinct components: high-frequency whine noise from the motor and background noise. This segmentation allows for separate analysis and prediction of each noise source, improving overall prediction accuracy by addressing specific noise characteristics rather than treating all noise uniformly.
Solution Approach 2:
The patent introduces an order analysis technique as an intermediary method to separate noise components based on their relationship with motor RPM. By using order analysis, the system can identify and isolate whine noise components that correlate with motor speed, enabling more accurate prediction without requiring complex physical measurements.
2Measurement precision
If noise sources are separated using order analysis, then the sound quality index prediction accuracy improves, but the evaluation process becomes more complex
Solution Approach 1:
The patent replaces complex physical noise measurement and separation systems with a data-driven approach using regression analysis and order analysis. Instead of using complex acoustic measurement equipment and manual separation techniques, the system uses computational methods to automatically separate and predict noise components, reducing measurement difficulty while maintaining accuracy.
3Reliability
If a comprehensive noise separation model is applied, then the NVH evaluation becomes more accurate, but the development time and computational resources increase
Solution Approach 1:
The patent performs preliminary order analysis and noise separation during the data collection phase, organizing noise data by its relationship with motor RPM before modeling. This preliminary organization of data makes subsequent regression analysis more efficient and reduces the computational time required during the development phase, as the data is already structured for analysis.
Solution Approach 2:
The patent transforms the noise prediction problem from a time-domain analysis to a frequency-domain analysis using order analysis, and then applies regression analysis on transformed parameters. By changing the analysis parameters and domains, the system achieves reliable NVH evaluation with reduced computational complexity and faster processing time.
Data Source
AI summary
A method of predicting a sound quality index of an electric vehicle, the method includes acquiring vehicle interior noise and vehicle data of the electric vehicle, evaluating, by a jury test, the sound quality index for a high-frequency whine noise component of the vehicle interior noise, extracting features to be used as predictors for modeling the sound quality index, learning a sound quality index model from the features to be used as predictors, and completing the sound quality index model, wherein the correlation of the sound quality index model with the sound quality index evaluation is 0.9 or more.