Active Control Algorithm Of Vehicle Interior Noise

Automotive NVH(noise,vibration and harshness)technology is one of the most important technologies to improve automobile quality.A key aspect of NVH research is interior noise control,which can be divided into two categories,namely passive noise control(PNC)and active noise control(ANC).Passive methods achieve the objective of noise reduction in automobile cabin by either modifying the structural properties,adding damping materials,or using some vibration isolation approaches.However,these methods are usually very effective from the mid-frequencies and upwards(higher than 500Hz)and ineffective for low frequency noise.Moreover,these approaches often require many design iterations in order to fulfill the desired sound target,which may result in more design cycle and long time.In addition,for automobile with different power sources,such as hybrid electric vehicle,passive noise control scheme is the tradeoff of the noise control methods between power sources and can not be switched adaptively according to the change of power source.The principle of ANC is acoustic superposition.ANC system introduces anti-noise,by integrating with car audio systems,to suppress automobile interior noise.Compared with PNC,ANC system is effective to control low frequency noise,which can not only be used to attenuate the sound pressure level(SPL)of the vehicle interior noise,but also be used to reshape the residual error spectrum to improve sound quality.The objective of conventional ANC system is to reduce the SPL of the error noise as much as possible.However,in automobile area,people not only require quiet,but also concern about the frequency components of the noise and the relationship between SPL of the interior noise and the engine speed.Moreover,the amplitude and frequency of the vehicle interior noise will be changed according to the driving condition.Hence,interior noise is nonstationary noise,which require fast convergence speed.Besides functionality,implementation cost of ANC system should be considered while new control algorithm is proposed in order to make ANC system suitable for mass production.This dissertation dedicates to solve the aforementioned technical problems,the main contents of the dissertation are as follows:Firstly,problem types of vehicle interior noise are summarized systematically.Noise problems which can be solved by active control technology are analyzed,which provides basis for active noise control algorithm research.“Source-transfer path-noise distribution” model is used to systematically summarize the sources of interior noise,transfer path,spatial distribution and frequency range.The applicability of ANC technology in vehicle interior noise control is discussed.Secondly,variable step size broadband active sound quality control(ASQC)algorithm is proposed to improve the convergence speed of the ASQC system.The proposed algorithm first normalize the filtered input signal by its power to assure the convergence of the algorithm in the presence of a time varying disturbance noise,and then adjust the step size dynamically according to the instantaneous pseudo error to make sure that ASQC system converge fast at the beginning of the control process and retain low mean square error(MSE)after convergence of the algorithm.Thirdly,Active broadband sound quality control algorithm with accurate predefined SPL is proposed to improve the linearity of interior noise SPL versus rpm,which reflects customers' desired perception on the feature of noise verses rpm.The proposed algorithm can not only shape the spectrum of the residual noise according to the sound quality filter,but also control the SPL of the residual noise to satisfy a predefined target accurately,when the amplitude of the disturbance noise changes.This aim is accomplished by introducing a gain factor which is adaptively determined based on the power of the instantaneous disturbance noise and the predefined target.Fourthly,hybrid active sound quality algorithm is proposed to reduce the cost of ASQC system.Feedforward control strategy is used to control the sound quality of narrowband noise whose frequency is related to engine speed.The amplitude of the noise can be attenuated or enhanced according to the requirement of interior noise target.Feedback control strategy is used to control broadband noise whose reference signal is difficult or expensive to acquire,such as road noise.Simplified hybrid ASQC algorithm is proposed to decrease the computational complexity.Finally,ANC experimental system is designed in the acoustic duct to validate the conventional and proposed ANC algorithm.The experimental system is designed according to the frequency range of the automobile interior noise.The primary path and secondary path is identified by least mean square(LMS)algorithm and is compared with the measurement results,which validate the accuracy of the identification.The identified secondary path is used in the design process of the ANC controller.Single frequency Sinusoidal noise and engine noise is used as primary noise in the experiment to validate ANC algorithm.And then the proposed variable step size broadband ASQC algorithm on validated in the experimental system.