Tire pattern noise simulation prediction method, device and program product based on image recognition

Abstract

The invention relates to the field of tire pattern intelligent design, in particular to a tire pattern noise simulation prediction method and device based on image recognition and a computer program product. A tread pattern noise simulation prediction method based on image recognition comprises the following steps: starting from an impact sounding mechanism of tread pattern noise, calculating a transverse sea-land ratio of a pattern, combining a running speed of a tire to obtain a time domain spectrum, and calculating a frequency spectrum through fast Fourier transform. The simulation principle is easy to master by pattern designers, the pretreatment operation is simple and rapid, the operation speed is high, the efficiency is high, the simulation result can meet the design analysis precision requirement, the new pattern research and development period is shortened, the development cost is reduced, and the pattern digital design level is improved.

Description

technical field [0001] The invention relates to the field of intelligent design of tire patterns, in particular to a method, equipment and computer program product for the simulation and prediction of tire pattern noise based on image recognition. Background technique [0002] Compared with fuel vehicles, the engine noise of electric vehicles is greatly reduced. The main source of vehicle noise is the engine being replaced by tires. With the popularity of electric vehicles, OEMs have higher and higher requirements for tire noise. The main test methods for tire noise are the field sliding method and the indoor drum method. The sliding method test requires a construction site, and the drum method requires the establishment of a noise laboratory. The construction cost is very expensive, and the test cost is high and the cycle is long. Therefore, major tire companies or research institutions are developing pattern noise simulation technology to guide pattern structure design and...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, the invention patent (publication number: CN104102820A publication date: 20141015) discloses a tire pattern noise analysis and prediction method, which is based on the sounding mechanism of the pump noise, and obtains the sounding characteristics of the pattern groove by identifying the pattern parameters, thereby calculating Tire pump noise, but this method needs to be calculated through the acoustic transfer function, which increases the uncertainty of the calculation results. The definition or expression of the acoustic transfer function is not mentioned in the patent. If the acoustic transfer function cannot accurately describe the groove The sound characteristics of the sound, resulting in a decrease in the accuracy of the prediction results, thereby degrading the performance of the design application

Invention patent application (publication number: CN113886991A publication date: 20220104) discloses a tire pattern impact noise prediction method, the method extracts the pattern design parameters by extracting points, straight lines and curves of the DXF format file through the program, and can only extract each In the case of the same rib pitch design, it is difficult to analyze when the rib pitch designs are different, and the frequency spectrum is calculated by Fourier series, and the frequency resolution is the fundamental frequency (if the outer circumference of the tire is Two meters, the fundamental frequency is 11.11Hz at a speed of 80km / h), so the frequency resolution is low

In addition, the invention patent application (publication number: CN112084682A publication date: 20201215) discloses a TBR tire noise test simulation method. This method uses large-scale analysis software for modeling and calculation analysis. The model is complex and time-consuming and labor-intensive. High requirements for modelers, not suitable for a large number of simulation calculations, and the cost of purchasing large-scale software is high

Images