| With the expansion of the automotive market,major automotive research and development centers are paying increasing attention to interior noise in order to improve vehicle driving comfort.The low-frequency noise inside the vehicle has strong penetrability and cannot be filtered out by traditional passive noise control methods,becoming the main factor affecting the vehicle’s NVH(Noise,Vibration,Harshness)performance.This article focuses on this issue and conducts in-depth research on active noise control technology.Combining digital signal processing technology,a noise active controller is designed using the Digital Signal Processor(DSP)and Code Composer Studio(CCS)software development platform.The feasibility of the controller is verified through experiments on active noise control in vehicles.Firstly,by analyzing the principle and structure of the active noise control system,it is determined that a feedforward control system is adopted in this paper,and a transverse finite impulse response(FIR)filter is selected for the filtering module inside the active controller.The least mean square algorithm is a representative of adaptive algorithms in the field of active control.In engineering practice,the optimized filter x least mean square algorithm is often used.This article derives the algorithm for the former and compares it with the latter,analyzing the negative effects of secondary channels in the control process.To address this issue,this product incorporates an offline identification process for acoustic,electrical,and physical pathways in the design process,utilizing the identified secondary channel model to eliminate or reduce its impact and improve the noise reduction effect of the controller in practical applications.The traditional fixed step adaptive algorithm lacks sufficient balance between convergence speed and steady-state error.Therefore,this paper proposes an optimized and improved VSSFx NLMS algorithm using symbolic functions,and compares it with the traditional Fx LMS algorithm through simulation analysis.This article uses a single channel active control algorithm to derive a dual channel active control algorithm,which uses two error sensors and two speakers for active noise control in space.Then,use Texas Instruments(TI)DSP chip TMS320F28335 for hardware selection and embedded software design.Referring to the product design scheme and performance indicators,AD7671 and DAC8552 were selected as AD/DA conversion modules.C language was used to develop embedded programs without operating systems on the CCS software development platform.The idea of modularization was used to write initialization programs for DSP and various chips,and communication protocols and data transmission between peripheral chips and active control chips were constructed,Using optimization and improvement algorithms to write offline identification programs for secondary channels and active control programs for dual channel systems.Finally,an experimental platform for active control of interior noise was built,and dual channel noise control experiments were conducted using the controller in the interior sound field environment.Noise reduction effects of about 10 d B(A)and about 5d B(A)were achieved in single frequency and dual frequency noise control experiments,respectively,verifying the functionality of the controller software system. |
