| As the vigorous development of the automobile industry today,to improve vehicle safety and economy is still the unswerving pursuit goal of each scientific research worker.The tire is one of the most important parts to guarantee normal driving of the car.The tire in the low-pressure condition for a long time,will increase tire wear,tire tear and fuel consumption,and will produce a very serious accident,especially in high-speed running state.Tire Pressure Monitoring System(TPMS)is the 3th largest security system after the anti-lock braking system and the airbag.The system can ensure vehicle safety in the level of tire failure,by monitoring the tire pressure state and giving alarm information to the driver.At present,TPMS mainly includes Direct TPMS and Indirect TPMS.Direct TPMS can get the tire pressure information directly through the pressure sensor.Comparatively,Indirect TPMS analyzes the wheel speed signal,which exists in ABS or ESP,to get the function of tire pressure monitoring and alarming.Indirect TPMS has advantage of low cost and convenient installation,which is beneficial for its application.Hence,Indirect TPMS has broad market prospect and far-reaching significance.Tire Pressure Estimation Algorithm based on wheel speed characteristic analysis is put forward in this thesis.The algorithm covers the tire pressure estimation algorithm based on frequency domain characteristics,the tire pressure estimation algorithm based on wheel rolling radius and the fusion algorithm of the two algorithms.Firstly,the experimental data is collected and the wheel speed characteristics are analyzed,and the original wheel speed signal is analyzed by wavelet theory in order to identify the interference factors and their causes.Secondly,the two-layer pretreatment of the wheel speed signal is carried out.The first layer preprocessing refers to the correction and homogenization of the wheel speed signal.The second layer preprocessing refers to the extraction of the effective and stable parts of the original signal.The correction of the wheel speed signal refers to the identification and correction of the ring gear error by the Recursive Least Squares(RLS)method.The wheel speed signal homogenization process means that the wheel speed signal is homogenized by linear interpolation.The high frequency component and the low frequency component of the wheel speed signal are extracted by filtering.Screening effective and stable signals requires the use of a variety of signal analysis methods to eliminate the non-stationary signal.Then,the frequency method based on FFT and AR model is researched,and radius statistic method is present based on the typical radius method,and the TPMS fusion algorithm based on the above two algorithms is studied.Finally,the tire pressure estimation algorithm model,based on the MATLAB/Simulink simulation platform,is established.The corresponding GUI interface is prepared.And the verification of the algorithm is implemented by real vehicle data. |
